https://ljefferson-menoko-site.netlify.app/assets/corpus.client-CBFtFu8N.js

1const e={caseStudyCount:8,writingCount:10,visibleProjectCount:8,builtAt:"2026-09-30T19:50:09.566Z"},t=[{slug:"gunkustom",project:"GunKustom",oneLiner:"A production commerce platform where a customer drags parts into a custom AR-platform build and the system guarantees they physically fit, which turned out to be a data-trust problem before it was ever a software problem.",status:"Production (live at gunkustom.com)",statusFamily:"production",tier:"primary",stack:["NestJS","TypeScript","Node.js","Express","Python","MongoDB / Mongoose","React","Vite","Tailwind CSS","Server-Sent Events","JWT / Google OAuth / TOTP 2FA","HMAC-SHA256 webhooks","Swagger / OpenAPI","Jest","Docker","Netlify","Render"],skills:["microservice architecture","idempotent ETL pipeline design","entity resolution and data-quality methodology","data modeling for computable constraints","human-in-the-loop classification","event-driven integration","server-sent events","authentication and authorization (JWT, OAuth, 2FA)","webhook signature verification","polite web scraping","observability and runbooks","technical leadership"],route:"/case-studies/gunkustom",image:"/case-studies/art/gunkustom.webp",links:[{label:"gunkustom.com",href:"https://gunkustom.com"}]},{slug:"powalert",project:"PowAlert",oneLiner:"A real-time snow-alert backend that turns resort forecasts and personal thresholds into one clean notification per user per cycle, correctly segmented for SMS, email and push, delivered exactly once.",status:"Production (live at powalert.com)",statusFamily:"production",tier:"primary",stack:["Node.js","Express","JavaScript (ESM)","MongoDB / Mongoose","node-cron","Twilio","SendGrid","Postmark","Expo Push","Visual Crossing API","Puppeteer","argon2","JWT","express-validator","Swagger / OpenAPI","Morgan","node-cache"],skills:["event-driven backend design","multi-channel notification architecture","Facade and Strategy patterns","greedy message segmentation","idempotent writes and exactly-once delivery","scheduled job design","third-party API integration","password hashing with argon2","JWT authentication","API documentation with Swagger","centralized error handling"],route:"/case-studies/powalert",image:"/case-studies/art/powalert.webp",links:[{label:"powalert.com",href:"https://powalert.com"}]},{slug:"ag3nt24",project:"Ag3nt24",oneLiner:"The open research harness I'm building in the open at classHuman AI, with inherited AI estates as a research track: scouts read an AI system a business can no longer explain, a Cedar policy boundary gates every crossing, and a human signs in HADES before anything changes state.",status:"Active development (design with early implementation, Apache-2.0)",statusFamily:"active",tier:"primary",stack:["Python 3.13","FastAPI","Pydantic","React","Vite","TypeScript","PostgreSQL","Docker compose","Cedar","MCP","uv","pytest","GitHub Actions"],skills:["AI governance and safety architecture","anti-corruption layer design","authorization policy design with Cedar","hash-chained append-only receipts","canonical serialization and evidence hashing","conformance pinning","architecture decision records","open-source project stewardship","legacy AI systems modernization"],route:"/case-studies/ag3nt24",image:"/case-studies/art/ag3nt24.webp",links:[{label:"Project site",href:"https://menokoog.github.io/Ag3nt24-oss"},{label:"Source (GitHub)",href:"https://github.com/MenokoOG/Ag3nt24-oss"}]},{slug:"learning-lounge",project:"M3n0koOg Learning Lounge",oneLiner:"An open-source classroom platform for free technical training, where the ingestion pipeline checks robots.txt and extracts the licence before it stores a word, and the paraphrase step falls back to a deterministic local rewriter when the model is unavailable.",status:"Active development (open source, MIT)",statusFamily:"active",tier:"primary",stack:["Node.js","TypeScript","Express","MongoDB / Mongoose","Bull + Redis (ioredis)","Puppeteer","Cheerio","@mozilla/readability","jsdom","OpenAI API","React","Vite","Tailwind CSS","Jest","Supertest"],skills:["accessibility-first interface design","content ingestion and normalization pipelines","queue-backed worker architecture","robots.txt compliance","license and copyright extraction","role-based access control","LLM integration with a deterministic fallback","REST API design","Jest and Supertest testing"],route:"/case-studies/learning-lounge",image:"/case-studies/art/learning-lounge.webp",links:[{label:"Source (GitHub)",href:"https://github.com/MenokoOG/M3n0koOg-Learning-Lounge"}]},{slug:"free-skills",project:"classHuman Free Skills",oneLiner:"Two free agent skills, a nine-field receipt convention, and a browser game about approving things, released together to make one argument: if a person never sets the last field, you do not have a human in the loop, you have a rubber stamp with extra steps.",status:"Shipped (public, free, no signup)",statusFamily:"production",tier:"primary",stack:["Markdown skill definitions","Python 3.10+ (no dependencies)","TypeScript / Node 18+ (no dependencies)","Single-file HTML/JS"],skills:["agent skill authoring","telemetry convention design","auditability and traceability","privacy-preserving instrumentation","reference implementation in two languages","game design as argument","technical writing","o

1pen-source release practice"],route:"/case-studies/free-skills",image:"/case-studies/art/free-skills.webp",links:[{label:"Source (GitHub)",href:"https://github.com/MenokoOG/free-skills"}]},{slug:"taco",project:"TACO Loop",oneLiner:"A decision-control architecture built on one inverted rule (unknown data must increase decision discipline, not model confidence) taken from concept to a published white paper and a formal mathematical model where the safety properties are theorems about the algebra rather than promises in prose.",status:"Research, White Paper v1.0 and formal model complete; implementation not yet public",statusFamily:"research",tier:"primary",stack:["Architecture and formal specification","Markdown","Python (target)","FastAPI (target)","PostgreSQL + pgvector (target)","Redis","Celery","React + Vite","React Native + Expo","OpenTelemetry"],skills:["systems and control architecture","AI safety and decision-control design","formal mathematical modeling","choosing algebraic form to encode a safety invariant","monotonicity and boundedness arguments","utility-based action selection under risk","guardrail AND-gate design with hard zero-invariants","hash-chained tamper-evident audit design","benchmark design with a falsifiable hypothesis","literature grounding and citation practice","technical writing","scoping non-claims as rigorously as claims"],route:"/case-studies/taco",image:"/case-studies/art/taco.webp"},{slug:"proforma",project:"ProForma",oneLiner:"A tool that turns a Gen AI idea into a 5-year c

1ost, benefit and risk projection, built for the person who has to defend the numbers in the room, and built so a refactor cannot quietly move one of those numbers.",status:"Shipped (public, Apache-2.0)",statusFamily:"production",tier:"primary",stack:["React 19","TypeScript","Vite","Vitest","hand-rolled SVG","localStorage","ESLint","Prettier","GitHub Actions"],skills:["verified calculation engines","regression testing against a source of truth","CI-enforced WCAG AA contrast testing","bundle-budget discipline","hand-rolled SVG data visualization","print-fidelity export","mobile-first responsive redesign","schema hydration and backward-compatible persistence","open-source licensing and attribution"],route:"/case-studies/proforma",image:"/case-studies/art/proforma.webp",links:[{label:"Live demo",href:"https://menokoog.github.io/proforma/"},{label:"Source (GitHub)",href:"https://github.com/MenokoOG/proforma"}]},{slug:"m3n0ko0g-website",project:"M3n0ko0g Website",oneLiner:"My portfolio rebuilt as an agent console, where a digital twin answers questions about my work by actually searching a build-time corpus in front of you, and the trace you watch is the retrieval that happened, not an animation of one.",status:"Production (live on Netlify)",statusFamily:"production",tier:"primary",stack:["React 18","TypeScript 5","Vite 6","Tailwind CSS 3","React Router 7","Netlify Functions","OpenAI API (server-side)","Vitest"],skills:["retrieval design without a vector database","BM25-style keyword and tag scoring","build-time corpus indexing","streaming LLM responses","graceful degradation to a deterministic search-only mode","serverless functions","secure server-side key handling","accessibility (reduced-motion, focus management, WCAG AA in two themes)","design-token discipline"],route:"/case-studies/m3n0ko0g-website",image:"/case-studies/art/m3n0ko0g-website.webp",links:[{label:"Live site",href:"https://ljefferson-menoko-site.netlify.app/"}]}],i=JSON.parse('[{"type":"case-study","project":"GunKustom","slug":"gunkustom","route":"/case-studies/gunkustom","statusFamily":"production","tier":"primary","stack":["NestJS","TypeScript","Node.js","Express","Python","MongoDB / Mongoose","React","Vite","Tailwind CSS","Server-Sent Events","JWT / Google OAuth / TOTP 2FA","HMAC-SHA256 webhooks","Swagger / OpenAPI","Jest","Docker","Netlify","Render"],"skills":["microservice architecture","idempotent ETL pipeline design","entity resolution and data-quality methodology","data modeling for computable constraints","human-in-the-loop classification","event-driven integration","server-sent events","authentication and authorization (JWT, OAuth, 2FA)","webhook signature verification","polite web scraping","observability and runbooks","technical leadership"],"sourceFile":"GunKustom.use-case.md","id":"gunkustom#summary","heading":"GunKustom","excerpt":"GunKustom. A production commerce platform where a customer drags parts into a custom AR-platform build and the system guarantees they physically fit, which turned out to be a data-trust problem before it was ever a software problem. Status…"},{"type":"case-study","project":"GunKustom","slug":"gunkustom","route":"/case-studies/gunkustom","statusFamily":"production","tier":"primary","stack":["NestJS","TypeScript","Node.js","Express","Python","MongoDB / Mongoose","React","Vite","Tailwind CSS","Server-Sent Events","JWT / Google OAuth / TOTP 2FA","HMAC-SHA256 webhooks","Swagger / OpenAPI","Jest","Docker","Netlify","Render"],"skills":["microservice architecture","idempotent ETL pipeline design","entity resolution and data-quality methodology","data modeling for computable constraints","human-in-the-loop classification","event-driven integration","server-sent events","authentication and authorization (JWT, OAuth, 2FA)","webhook signature verification","polite web scraping","observability and runbooks","technical leadership"],"sourceFile":"GunKustom.use-case.md","id":"gunkustom#problem","heading":"Problem","excerpt":"A customer assembling a custom AR-platform firearm is making a mechanical decision, not a shopping decision. Gas 

1system lengths have to match. Barrel nut types have to match the handguard. Receivers have to be within mil-spec tolerance of …"},{"type":"case-study","project":"GunKustom","slug":"gunkustom","route":"/case-studies/gunkustom","statusFamily":"production","tier":"primary","stack":["NestJS","TypeScript","Node.js","Express","Python","MongoDB / Mongoose","React","Vite","Tailwind CSS","Server-Sent Events","JWT / Google OAuth / TOTP 2FA","HMAC-SHA256 webhooks","Swagger / OpenAPI","Jest","Docker","Netlify","Render"],"skills":["microservice architecture","idempotent ETL pipeline design","entity resolution and data-quality methodology","data modeling for computable constraints","human-in-the-loop classification","event-driven integration","server-sent events","authentication and authorization (JWT, OAuth, 2FA)","webhook signature verification","polite web scraping","observability and runbooks","technical leadership"],"sourceFile":"GunKustom.use-case.md","id":"gunkustom#what-i-built","heading":"What I Built","excerpt":"I joined as Senior Lead Back-End Engineer and was promoted to CTO about six months in. As CTO I led a ground-up re-architecture from a monolith toward a set of focused services with clear ownership boundaries, working alongside a senior en…"},{"type":"case-study","project":"GunKustom","slug":"gunkustom","route":"/case-studies/gunkustom","statusFamily":"production","tier":"primary","stack":["NestJS","TypeScript","Node.js","Express","Python","MongoDB / Mongoose","React","Vite","Tailwind CSS","Server-Sent Events","JWT / Google OAuth / TOTP 2FA","HMAC-SHA256 webhooks","Swagger / OpenAPI","Jest","Docker","Netlify","Render"],"skills":["microservice architecture","idempotent ETL pipeline design","entity resolution and data-quality methodology","data modeling for computable constraints","human-in-the-loop classification","event-driven integration","server-sent events","authentication and authorization (JWT, OAuth, 2FA)","webhook signature verification","polite web scraping","observability and runbooks","technical leadership"],"sourceFile":"GunKustom.use-case.md","id":"gunkustom#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"Microservice architecture and service-boundary design, NestJS, TypeScript, Node.js, Express, Python, MongoDB and Mongoose, idempotent ETL pipeline design, bulk-write batching and bounded-memory data processing, entity resolution and data-q…"},{"type":"case-study","project":"GunKustom","slug":"gunkustom","route":"/case-studies/gunkustom","statusFamily":"production","tier":"primary","stack":["NestJS","TypeScript","Node.js","Express","Python","MongoDB / Mongoose","React","Vite","Tailwind CSS","Server-Sent Events","JWT / Google OAuth / TOTP 2FA","HMAC-SHA256 webhooks","Swagger / OpenAPI","Jest","Docker","Netlify","Render"],"skills":["microservice architecture","idempotent ETL pipeline design","entity resolution and data-quality methodology","data modeling for computable constraints","human-in-the-loop classification","event-driven integration","server-sent events","authentication and authorization (JWT, OAuth, 2FA)","webhook signature verification","polite web scraping","observability and runbooks","technical leadership"],"sourceFile":"GunKustom.use-case.md","id":"gunkustom#outcome-evidence","heading":"Outcome / Evidence","excerpt":"The platform shipped to production at gunkustom.com. The re-architecture I led took GunKustom from a single monolith to a maintainable set of services with clear boundaries, a data pipeline that could be re-run with confidence, and a catal…"},{"type":"case-study","project":"GunKustom","slug":"gunkustom","route":"/case-studies/gunkustom","statusFamily":"production","tier":"primary","stack":["NestJS","TypeScript","Node.js","Express","Python","MongoDB / Mongoose","React","Vite","Tailwind CSS","Server-Sent Events","JWT / Google OAuth / TOTP 2FA","HMAC-SHA256 webhooks","Swagger / OpenAPI","Jest","Docker","Netlify","Render"],"skills":["microservice architecture","idempotent ETL pipeline design","entity resolution and data-quality methodology","data modeling for computable constraints","human-in-the-loop classification","event-driven integration","server-sent events","authentication and authorization (JWT, OAuth, 2FA)","webhook signature verification","polite web scraping","observability and runbooks","technical leadership"],"sourceFile":"GunKustom.use-case.md","id":"gunkustom#interview-talking-points","heading":"Interview Talking Points","excerpt":"The compatibility promise is a data-trust promise. The collection spec mandates barcode structure validation and cross-checking every part against a second source, with discrepancies flagged in a notes field rather than silently resolved. …"},{"type":"case-study","project":"PowAlert","slug":"powalert","route":"/case-studies/powalert","statusFamily":"production","tier":"primary","stack":["Node.js","Express","JavaScript (ESM)","MongoDB / Mongoose","node-cron","Twilio","SendGrid","Postmark","Expo Push","Visual Crossing API","Puppeteer","argon2","JWT","express-validator","Swagger / OpenAPI","Morgan","node-cache"],"skills":["event-driven backend design","multi-channel notification architecture","Facade and Strategy patterns","greedy message segmentation","idempotent writes and exactly-once delivery","scheduled job design","third-party API integration","password hashing with argon2","JWT authentication","API documentation with Swagger","centralized error handling"],"sourceFile":"powalert.use-case.md","id":"powalert#summary","heading":"PowAlert","excerpt":"PowAlert. A real-time snow-alert backend that turns resort forecasts and personal thresholds into one clean notification per user per cycle, correctly segmented for SMS, email and push, delivered exactly once. Status: Production (live at p…"},{"type":"case-study","project":"PowAlert","slug":"powalert","route":"/case-studies/powalert","statusFamily":"production","tier":"primary","stack":["Node.js","Express","JavaScript (ESM)","MongoDB / Mongoose","node-cron","Twilio","SendGrid","Postmark","Expo Push","Visual Crossing API","Puppeteer","argon2","JWT","express-validator","Swagger / OpenAPI","Morgan","node-cache"],"skills":["event-driven backend design","multi-channel notification architecture","Facade and Strategy patterns","greedy message segmentation","idempotent writes and exactly-once delivery","scheduled job design","third-party API integration","password hashing with argon2","JWT authentication","API documentation with Swagger","centralized error handling"],"sourceFile":"powalert.use-case.md","id":"powalert#problem","heading":"Problem","excerpt":"PowAlert notifies skiers when conditions at the resorts they care about cross the thresholds they set (snowfall, snow depth, storm timing) and links them straight to lodging so they can act on it. The entire product is the notification. Ev…"},{"type":"case-study","project":"PowAlert","slug":"powalert","route":"/case-studies/powalert","statusFamily":"production","tier":"primary","stack":["Node.js","Express","JavaScript (ESM)","MongoDB / Mongoose","node-cron","Twilio","SendGrid","Postmark","Expo Push","Visual Crossing API","Puppeteer","argon2","JWT","express-validator","Swagger / OpenAPI","Morgan","node-cache"],"skills":["event-driven backend design","multi-channel notification architecture","Facade and Strategy patterns","greedy message segmentation","idempotent writes and exactly-once delivery","scheduled job design","third-party API integration","password hashing with argon2","JWT authentication","API documentation with Swagger","centralized error handling"],"sourceFile":"powalert.use-case.md","id":"powalert#what-i-built","heading":"What I Built","excerpt":"I led the design and rebuild of the PowAlert backend (an 

1Express service on Node) owning its architecture, the multi-channel notification engine, the alert-processing algorithms, and the scheduled ingestion jobs. The brief was to turn ad-h…"},{"type":"case-study","project":"PowAlert","slug":"powalert","route":"/case-studies/powalert","statusFamily":"production","tier":"primary","stack":["Node.js","Express","JavaScript (ESM)","MongoDB / Mongoose","node-cron","Twilio","SendGrid","Postmark","Expo Push","Visual Crossing API","Puppeteer","argon2","JWT","express-validator","Swagger / OpenAPI","Morgan","node-cache"],"skills":["event-driven backend design","multi-channel notification architecture","Facade and Strategy patterns","greedy message segmentation","idempotent writes and exactly-once delivery","scheduled job design","third-party API integration","password hashing with argon2","JWT authentication","API documentation with Swagger","centralized error handling"],"sourceFile":"powalert.use-case.md","id":"powalert#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"Event-driven backend design, Node.js, Express, JavaScript ESM, MongoDB and Mongoose, multi-channel notification architecture, Facade and Strategy patterns applied where change actually happens, message aggregation algorithms, greedy segmen…"},{"type":"case-study","project":"PowAlert","slug":"powalert","route":"/case-studies/powalert","statusFamily":"production","tier":"primary","stack":["Node.js","Express","JavaScript (ESM)","MongoDB / Mongoose","node-cron","Twilio","SendGrid","Postmark","Expo Push","Visual Crossing API","Puppeteer","argon2","JWT","express-validator","Swagger / OpenAPI","Morgan","node-cache"],"skills":["event-driven backend design","multi-channel notification architecture","Facade and Strategy patterns","greedy message segmentation","idempotent writes and exactly-once delivery","scheduled job design","third-party API integration","password hashing with argon2","JWT authentication","API documentation with Swagger","centralized error handling"],"sourceFile":"powalert.use-case.md","id":"powalert#outcome-evidence","heading":"Outcome / Evidence","excerpt":"The rebuilt backend shipped as the engine behind PowAlert\'s launched MVP, live at powalert.com. The pattern-driven design paid off exactly where it was aimed: the notification system became a small, well-factored core that absorbed a third…"},{"type":"case-study","project":"PowAlert","slug":"powalert","route":"/case-studies/powalert","statusFamily":"production","tier":"primary","stack":["Node.js","Express","JavaScript (ESM)","MongoDB / Mongoose","node-cron","Twilio","SendGrid","Postmark","Expo Push","Visual Crossing API","Puppeteer","argon2","JWT","express-validator","Swagger / OpenAPI","Morgan","node-cache"],"skills":["event-driven backend design","multi-channel notification architecture","Facade and Strategy patterns","greedy message segmentation","idempotent writes and exactly-once delivery","scheduled job design","third-party API integration","password hashing with argon2","JWT authentication","API documentation with Swagger","centralized error handling"],"sourceFile":"powalert.use-case.md","id":"powalert#interview-talking-points","heading":"Interview Talking Points","excerpt":"The design attention went to the component most likely to change. Delivery channels churn; alert logic does not. A facade over delivery plus a per-channel segmentation strategy means adding push notifications was additive rather than invas…"},{"type":"case-study","project":"Ag3nt24","slug":"ag3nt24","route":"/case-studies/ag3nt24","statusFamily":"active","tier":"primary","stack":["Python 3.13","FastAPI","Pydantic","React","Vite","TypeScript","PostgreSQL","Docker compose","Cedar","MCP","uv","pytest","GitHub Actions"],"skills":["AI governance and safety architecture","anti-corruption layer design","authorization policy design with Cedar","hash-chained append-only receipts","canonical serialization and evidence hashing","conformance pinning","architecture decision records","open-source project stewardship","legacy AI systems modernization"],"sourceFile":"Ag3nt24.use-case.md","id":"ag3nt24#summary","heading":"Ag3nt24","excerpt":"Ag3nt24. The open research harness I\'m building in the open at classHuman AI, with inherited AI estates as a research track: scouts read an AI system a business can no longer explain, a Cedar policy boun

1dary gates every crossing, and a hum…"},{"type":"case-study","project":"Ag3nt24","slug":"ag3nt24","route":"/case-studies/ag3nt24","statusFamily":"active","tier":"primary","stack":["Python 3.13","FastAPI","Pydantic","React","Vite","TypeScript","PostgreSQL","Docker compose","Cedar","MCP","uv","pytest","GitHub Actions"],"skills":["AI governance and safety architecture","anti-corruption layer design","authorization policy design with Cedar","hash-chained append-only receipts","canonical serialization and evidence hashing","conformance pinning","architecture decision records","open-source project stewardship","legacy AI systems modernization"],"sourceFile":"Ag3nt24.use-case.md","id":"ag3nt24#problem","heading":"Problem","excerpt":"A business has an AI system it can no longer explain. Prompt chains, retrieval pipelines, fine-tuned models, vector stores and orchestration glue, bolted together between 2021 and 2026 while the field moved underneath them. The people who …"},{"type":"case-study","project":"Ag3nt24","slug":"ag3nt24","route":"/case-studies/ag3nt24","statusFamily":"active","tier":"primary","stack":["Python 3.13","FastAPI","Pydantic","React","Vite","TypeScript","PostgreSQL","Docker compose","Cedar","MCP","uv","pytest","GitHub Actions"],"skills":["AI governance and safety architecture","anti-corruption layer design","authorization policy design with Cedar","hash-chained append-only receipts","canonical serialization and evidence hashing","conformance pinning","architecture decision records","open-source project stewardship","legacy AI systems modernization"],"sourceFile":"Ag3nt24.use-case.md","id":"ag3nt24#what-i-built","heading":"What I Built","excerpt":"Ag3nt24 is open source under Apache-2.0 at MenokoOG/Ag3nt24-oss. I\'m CEO, CTO and lead architect at classHuman AI LLC, and this is the project I build in public. What exists today. The design, 18 architecture decision records, the contract…"},{"type":"case-study","project":"Ag3nt24","slug":"ag3nt24","route":"/case-studies/ag3nt24","statusFamily":"active","tier":"primary","stack":["Python 3.13","FastAPI","Pydantic","React","Vite","TypeScript","PostgreSQL","Docker compose","Cedar","MCP","uv","pytest","GitHub Actions"],"skills":["AI governance and safety architecture","anti-corruption layer design","authorization policy design with Cedar","hash-chained append-only receipts","canonical serialization and evidence hashing","conformance pinning","architecture decision records","open-source project stewardship","legacy AI systems modernization"],"sourceFile":"Ag3nt24.use-case.md","id":"ag3nt24#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"AI governance and safety architecture, anti-corruption layer design, authorization policy design with Cedar, Ed25519 signing and verified principals, hash-chained append-only receipts, canonical serialization and evidence hashing, Pydantic…"},{"type":"case-study","project":"Ag3nt24","slug":"ag3nt24","route":"/case-studies/ag3nt24","statusFamily":"active","tier":"primary","stack":["Python 3.13","FastAPI","Pydantic","React","Vite","TypeScript","PostgreSQL","Docker compose","Cedar","MCP","uv","pytest","GitHub Actions"],"skills":["AI governance and safety architecture","anti-corruption layer design","authorization policy design with Cedar","hash-chained append-only receipts","canonical serialization and evidence hashing","conformance pinning","architecture decision records","open-source project stewardship","legacy AI systems modernization"],"sourceFile":"Ag3nt24.use-case.md","id":"ag3nt24#outcome-evidence","heading":"Outcome / Evidence","excerpt":"Public Apache-2.0 repository with CI at github.com/MenokoOG/Ag3nt24-oss, project site at menokoog.github.io/Ag3nt24-oss. Measured on 2026-09-14 against `main`: the contracts package runs 28 tests, all passing (`uv sync && uv run pytest`). …"},{"type":"case-study","project":"Ag3nt24","slug":"ag3nt24","route":"/case-studies/ag3nt24","statusFamily":"active","tier":"primary","stack":["Python 3.13","FastAPI","Pydantic","React","Vite","TypeScript","PostgreSQL","Docker compose","Cedar","MCP","uv","pytest","GitHub Actions"],"skills":["AI governance and safety architecture","anti-corruption layer design","authorization policy design with Cedar","hash-chained append-only receipts","canonical serialization and evidence hashing","conformance pinning","architecture decision records","open-source project stewardship","legacy AI systems modernization"],"sourceFile":"Ag3nt24.use-case.md","id":"ag3nt24#interview-talking-points","heading":"Interview Talking Points","excerpt":"\\"Intelligence is flexible. Authority is stable.\\" The layer that decides what an agent may do gives the same verdict for the same input. I picked Cedar for that: default deny, no I/O, and a policy change can be proven not to widen access. T…"},{"type":"case-study","project":"M3n0koOg Learning Lounge","slug":"learning-lounge","route":"/case-studies/learning-lounge","statusFamily":"active","tier":"primary","stack":["Node.js","TypeScript","Express","MongoDB / Mongoose","Bull + Redis (ioredis)","Puppeteer","Cheerio","@mozilla/readability","jsdom","OpenAI API","React","Vite","Tailwind CSS","Jest","Supertest"],"skills":["accessibility-first interface design","content ingestion and normalization pipelines","queue-backed worker architecture","robots.txt compliance","license and copyright extraction","role-based access control","LLM integration with a deterministic fallback","REST API design","Jest and Supertest testing"],"sourceFile":"learning-lounge.use-case.md","id":"learning-lounge#summary","heading":"M3n0koOg Learning Lounge","excerpt":"M3n0koOg Learning Lounge. An open-source classroom platform for free technical training, where the ingestion pipeline checks robots.txt and extracts the licence before it stores a word, and the paraphrase step falls back to a deterministic…"},{"type":"case-study","project":"M3n0koOg Learning Lounge","slug":"learning-lounge","route":"/case-studies/learning-lounge","statusFamily":"active","tier":"primary","stack":["Node.js","TypeScript","Express","MongoDB / Mongoose","Bull + Redis (ioredis)","Puppeteer","Cheerio","@mozilla/readability","jsdom","OpenAI API","React","Vite","Tailwind CSS","Jest","Supertest"],"skills":["accessibility-first interface design","content ingestion and normalization pipelines","queue-backed worker architecture","robots.txt compliance","license and copyright extraction","role-based access control","LLM integration with a deterministic fallback","REST API design","Jest and Supertest testing"],"sourceFile":"learning-lounge.use-case.md","id":"learning-lounge#problem","heading":"Problem","excerpt":"Good free technical training exists. It is just unusable for a lot of the people who need it most. The material is scattered across dozens of sites with incompatible navigation. It is written for someone who already reads well, in low-cont…"},{"type":"case-study","project":"M3n0koOg Learning Lounge","slug":"learning-lounge","route":"/case-studies/learning-lounge","statusFamily":"active","tier":"primary","stack":["Node.js","TypeScript","Express","MongoDB / Mongoose","Bull + Redis (ioredis)","Puppeteer","Cheerio","@mozilla/readability","jsdom","OpenAI API","React","Vite","Tailwind CSS","Jest","Supertest"],"skills":["accessibility-first interface design","content ingestion and normalization pipelines","queue-backed worker architecture","robots.txt compliance","license and copyright extraction","role-based access control","LLM integration with a deterministic fallback","REST API design","Jest and Supertest testing"],"sourceFile":"learning-lounge.use-case.md","id":"learning-lounge#what-i-built","heading":"What I Built","excerpt":"An open-source full-stack platform under MIT, with a mentorship-oriented contribution model, issues labelled `mentored` so students have a marked path into a real codebase rather than being told to \\"find a good first issue.\\" A Node and Typ…"},{"type":"case-study","project":"M3n0koOg Learning Lounge","slug":"learning-lounge","route":"/case-studies/learning-lounge","statusFamily":"active","tier":"primary","stack":["Node.js","TypeScript","Express","MongoDB / Mongoose","Bull + Redis (ioredis)","Puppeteer","Cheerio","@mozilla/readability","jsdom","OpenAI API","React","Vite","Tailwind CSS","Jest","Supertest"],"skills":["accessibility-first interface design","content ingestion and normalization pipelines","queue-backed worker architecture","robots.txt compliance","license and copyright extraction","role-based access control","LLM integration with a deterministic fallback","REST API design","Jest and Supertest testing"],"sourceFile":"learning-lounge.use-case.md","id":"learning-lounge#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"Accessibility-first product thinking, Node.js, TypeScript, Express, MongoDB and Mongoose schema design, queue-backed worker architecture with Bull and Redis, out-of-process job isolation, web scraping with Cheerio and Puppeteer, article ex…"},{"type":"case-study","project":"M3n0koOg Learning Lounge","slug":"learning-lounge","route":"/case-studies/learning-lounge","statusFamily":"active","tier":"primary","stack":["Node.js","TypeScript","Express","MongoDB / Mongoose","Bull + Redis (ioredis)","Puppeteer","Cheerio","@mozilla/readability","jsdom","OpenAI API","React","Vite","Tailwind CSS","Jest","Supertest"],"skills":["accessibility-first interface design","content ingestion and normalization pipelines","queue-backed worker architecture","robots.txt compliance","license and copyright extraction","role-based access control","LLM integration with a deterministic fallback","REST API design","Jest and Supertest testing"],"sourceFile":"learning-lounge.use-case.md","id":"learning-lounge#outcome-evidence","heading":"Outcome / Evidence","excerpt":"Open-source under MIT at github.com/MenokoOG/M3n0koOg-Learning-Lounge, with contribution and code-of-conduct docs, a security policy, issue and PR templates, and `mentored` labels for student contributors. Shipped and verifiable in the rep…"},{"type":"case-study","project":"M3n0koOg Learning Lounge","slug":"learning-lounge","route":"/case-studies/learning-lounge","statusFamily":"active","tier":"primary","stack":["Node.js","TypeScript","Express","MongoDB / Mongoose","Bull + Redis (ioredis)","Puppeteer","Cheerio","@mozilla/readability","jsdom","OpenAI API","React","Vite","Tailwind CSS","Jest","Supertest"],"skills":["accessibility-first interface design","content ingestion and normalization pipelines","queue-backed worker architecture","robots.txt compliance","license and copyright extraction","role-based access control","LLM integration with a deterministic fallback","REST API design","Jest and Supertest testing"],"sourceFile":"learning-lounge.use-case.md","id":"learning-lounge#interview-talking-points","heading":"Interview Talking Points","excerpt":"Ingestion ethics belong in the pipeline, not in a policy document. `robots_check.ts` gates the fetch and `extract_license.ts` runs on every capture, so compliance happens on every ingest instead of whenever someone remembers. A paragraph o…"},{"type":"case-study","project":"classHuman Free Skills","slug":"free-skills","route":"/case-studies/free-skills","statusFamily":"production","tier":"primary","stack":["Mar

1kdown skill definitions","Python 3.10+ (no dependencies)","TypeScript / Node 18+ (no dependencies)","Single-file HTML/JS"],"skills":["agent skill authoring","telemetry convention design","auditability and traceability","privacy-preserving instrumentation","reference implementation in two languages","game design as argument","technical writing","open-source release practice"],"sourceFile":"free-skills.use-case.md","id":"free-skills#summary","heading":"classHuman Free Skills","excerpt":"classHuman Free Skills. Two free agent skills, a nine-field receipt convention, and a browser game about approving things, released together to make one argument: if a person never sets the last field, you do not have a human in the loop, …"},{"type":"case-study","project":"classHuman Free Skills","slug":"free-skills","route":"/case-studies/free-skills","statusFamily":"production","tier":"primary","stack":["Markdown skill definitions","Python 3.10+ (no dependencies)","TypeScript / Node 18+ (no dependencies)","Single-file HTML/JS"],"skills":["agent skill authoring","telemetry convention design","auditability and traceability","privacy-preserving instrumentation","reference implementation in two languages","game design as argument","technical writing","open-source release practice"],"sourceFile":"free-skills.use-case.md","id":"free-skills#problem","heading":"Problem","excerpt":"Teams are running agent skills in production right now and cannot answer three basic questions about them: how often did that run, what did it produce, and did anyone look at the output. The work happens. Something appears. Everyone moves …"},{"type":"case-study","project":"classHuman Free Skills","slug":"free-skills","route":"/case-studies/free-skills","statusFamily":"production","tier":"primary","stack":["Markdown skill definitions","Python 3.10+ (no dependencies)","TypeScript / Node 18+ (no dependencies)","Single-file HTML/JS"],"skills":["agent skill authoring","telemetry convention design","auditability and traceability","privacy-preserving instrumentation","reference implementation in two languages","game design as argument","technical writing","open-source release practice"],"sourceFile":"free-skills.use-case.md","id":"free-skills#what-i-built","heading":"What I Built","excerpt":"A small public repository (free, no signup, no email wall) containing two skills, a telemetry convention with reference implementations, a hand-maintained registry, and a game. Each piece is separately useful; together they are a position …"},{"type":"case-study","project":"classHuman Free Skills","slug":"free-skills","route":"/case-studies/free-skills","statusFamily":"production","tier":"primary","stack":["Markdown skill definitions","Python 3.10+ (no dependencies)","TypeScript / Node 18+ (no dependencies)","Single-file HTML/JS"],"skills":["agent skill authoring","telemetry convention design","auditability and traceability","privacy-preserving instrumentation","reference implementation in two languages","game design as argument","technical writing","open-source release practice"],"sourceFile":"free-skills.use-case.md","id":"free-skills#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"Agent skill authoring and prompt-as-specification design, telemetry convention design, auditability and traceability engineering, privacy-preserving instrumentation, append-only log design, identifier design and the timestamp-as-identity f…"},{"type":"case-study","project":"classHuman Free Skills","slug":"free-skills","route":"/case-studies/free-skills","statusFamily":"production","tier":"primary","stack":["Markdown skill definitions","Python 3.10+ (no dependencies)","TypeScript / Node 18+ (no dependencies)","Single-file HTML/JS"],"skills":["agent skill authoring","telemetry convention design","auditability and traceability","privacy-preserving instrumentation","reference implementation in two languages","game design as argument","technical writing","open-source release practice"],"sourceFile":"free-skills.use-case.md","id":"free-skills#outcome-evidence","heading":"Outcome / Evidence","excerpt":"Public and free at github.com/MenokoOG/free-skills, no signup, no email wall, released by classHuman AI LLC for public use with attribution appreciated but not required. The skills are plain Markdown with frontmatter and no runtime depende…"},{"type":"case-study","project":"classHuman Free Skills","slug":"free-skills","route":"/case-studies/free-skills","statusFamily":"production","tier":"primary","stack":["Mar

1kdown skill definitions","Python 3.10+ (no dependencies)","TypeScript / Node 18+ (no dependencies)","Single-file HTML/JS"],"skills":["agent skill authoring","telemetry convention design","auditability and traceability","privacy-preserving instrumentation","reference implementation in two languages","game design as argument","technical writing","open-source release practice"],"sourceFile":"free-skills.use-case.md","id":"free-skills#interview-talking-points","heading":"Interview Talking Points","excerpt":"The whole repository is one argument in three forms, a field, a convention, and a game. `human` starts at `pending` and only a person changes it; `grep -c \'\\"human\\":\\"pending\\"\' receipts.jsonl` tells you whether your review process is real; a…"},{"type":"case-study","project":"TACO Loop","slug":"taco","route":"/case-studies/taco","statusFamily":"research","tier":"primary","stack":["Architecture and formal specification","Markdown","Python (target)","FastAPI (target)","PostgreSQL + pgvector (target)","Redis","Celery","React + Vite","React Native + Expo","OpenTelemetry"],"skills":["systems and control architecture","AI safety and decision-control design","formal mathematical modeling","choosing algebraic form to encode a safety invariant","monotonicity and boundedness arguments","utility-based action selection under risk","guardrail AND-gate design with hard zero-invariants","hash-chained tamper-evident audit design","benchmark design with a falsifiable hypothesis","literature grounding and citation practice","technical writing","scoping non-claims as rigorously as claims"],"sourceFile":"taco.use-case.md","id":"taco#summary","heading":"TACO Loop","excerpt":"TACO Loop. A decision-control architecture built on one inverted rule (unknown data must increase decision discipline, not model confidence) taken from concept to a published white paper and a formal mathematical model where the safety pro…"},{"type":"case-study","project":"TACO Loop","slug":"taco","route":"/case-studies/taco","statusFamily":"research","tier":"primary","stack":["Architecture and formal specification","Markdown","Python (target)","FastAPI (target)","PostgreSQL + pgvector (target)","Redis","Celery","React + Vite","React Native + Expo","OpenTelemetry"],"skills":["systems and control architecture","AI safety and decision-control design","formal mathematical modeling","choosing algebraic form to encode a safety invariant","monotonicity and boundedness arguments","utility-based action selection under risk","guardrail AND-gate design with hard zero-invariants","hash-chained tamper-evident audit design","benchmark design with a falsifiable hypothesis","literature grounding and citation practice","technical writing","scoping non-claims as rigorously as claims"],"sourceFile":"taco.use-case.md","id":"taco#problem","heading":"Problem","excerpt":"Most AI and human workflow systems behave as though the input in front of them is sufficient to decide with. Real environments do not work that way. Operational decisions routinely start from data that is incomplete, contradictory, stale, …"},{"type":"case-study","project":"TACO Loop","slug":"taco","route":"/case-studies/taco","statusFamily":"research","tier":"primary","stack":["Architecture and formal specification","Markdown","Python (target)","FastAPI (target)","PostgreSQL + pgvector (target)","Redis","Celery","React + Vite","React Native + Expo","OpenTelemetry"],"skills":["systems and control architecture","AI safety and decision-control design","formal mathematical modeling","choosing algebraic form to encode a safety invariant","monotonicity and boundedness arguments","utility-based action selection under risk","guardrail AND-gate design with hard zero-invariants","hash-chained tamper-evident audit design","benchmark design with a falsifiable hypothesis","literature grounding and citation practice","technical writing","scoping non-claims as rigorously as claims"],"sourceFile":"taco.use-case.md","id":"taco#what-i-built","heading":"What I Built","excerpt":"The loop, and the test attached to each stage. TACO is Take In Unknowns → Assess and Align → Choose Correctly → Operate and Observe Outcome. What makes it a control architecture rather than a mnemonic is that each stage c

1arries a pass/fail…"},{"type":"case-study","project":"TACO Loop","slug":"taco","route":"/case-studies/taco","statusFamily":"research","tier":"primary","stack":["Architecture and formal specification","Markdown","Python (target)","FastAPI (target)","PostgreSQL + pgvector (target)","Redis","Celery","React + Vite","React Native + Expo","OpenTelemetry"],"skills":["systems and control architecture","AI safety and decision-control design","formal mathematical modeling","choosing algebraic form to encode a safety invariant","monotonicity and boundedness arguments","utility-based action selection under risk","guardrail AND-gate design with hard zero-invariants","hash-chained tamper-evident audit design","benchmark design with a falsifiable hypothesis","literature grounding and citation practice","technical writing","scoping non-claims as rigorously as claims"],"sourceFile":"taco.use-case.md","id":"taco#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"Systems and control architecture, AI safety and decision-control design, formal mathematical modeling, choosing algebraic form to encode a safety invariant, monotonicity and boundedness arguments, normalization and comparability design, ut…"},{"type":"case-study","project":"TACO Loop","slug":"taco","route":"/case-studies/taco","statusFamily":"research","tier":"primary","stack":["Architecture and formal specification","Markdown","Python (target)","FastAPI (target)","PostgreSQL + pgvector (target)","Redis","Celery","React + Vite","React Native + Expo","OpenTelemetry"],"skills":["systems and control architecture","AI safety and decision-control design","formal mathematical modeling","choosing algebraic form to encode a safety invariant","monotonicity and boundedness arguments","utility-based action selection under risk","guardrail AND-gate design with hard zero-invariants","hash-chained tamper-evident audit design","benchmark design with a falsifiable hypothesis","literature grounding and citation practice","technical writing","scoping non-claims as rigorously as claims"],"sourceFile":"taco.use-case.md","id":"taco#outcome-evidence","heading":"Outcome / Evidence","excerpt":"White Paper v1.0, twelve sections plus an MVP backlog appendix: executive summary, problem statement, the TACO definition with per-stage operational tests, source-validated foundations against ten citations, the mathematical model, the twe…"},{"type":"case-study","project":"TACO Loop","slug":"taco","route":"/case-studies/taco","statusFamily":"research","tier":"primary","stack":["Architecture and formal specification","Markdown","Python (target)","FastAPI (target)","PostgreSQL + pgvector (target)","Redis","Celery","React + Vite","React Native + Expo","OpenTelemetry"],"skills":["systems and control architecture","AI safety and decision-control design","formal mathematical modeling","choosing algebraic form to encode a safety invariant","monotonicity and boundedness arguments","utility-based action selection under risk","guardrail AND-gate design with hard zero-invariants","hash-chained tamper-evident audit design","benchmark design with a falsifiable hypothesis","literature grounding and citation practice","technical writing","scoping non-claims as rigorously as claims"],"sourceFile":"taco.use-case.md","id":"taco#interview-talking-points","heading":"Interview Talking Points","excerpt":"The failure mode I built around is not that models are overconfident, it is that thin evidence is invisible to them. A model handed less produces the same fluent output, and fluency reads as certainty downstream. Thin evidence has to act a…"},{"type":"case-study","project":"ProForma","slug":"proforma","route":"/case-studies/proforma","statusFamily":"production","tier":"primary","stack":["React 19","TypeScript","Vite","Vitest","hand-rolled SVG","localStorage","ESLint","Prettier","GitHub Actions"],"skills":["verified calculation engines","regression testing against a source of truth","CI-enforced WCAG AA contrast testing","bundle-budget discipline","hand-rolled SVG data visualization","print-fidelity export","mobile-first responsive redesign","schema hydration and backward-compatible persistence","open-source licensing and attribution"],"sourceFile":"proforma.use-case.md","id":"proforma#summary","heading":"ProForma","excerpt":"ProForma. A tool that turns a Gen AI idea into a 5-year c

1ost, benefit and risk projection, built for the person who has to defend the numbers in the room, and built so a refactor cannot quietly move one of those numbers. Status: Shipped (p…"},{"type":"case-study","project":"ProForma","slug":"proforma","route":"/case-studies/proforma","statusFamily":"production","tier":"primary","stack":["React 19","TypeScript","Vite","Vitest","hand-rolled SVG","localStorage","ESLint","Prettier","GitHub Actions"],"skills":["verified calculation engines","regression testing against a source of truth","CI-enforced WCAG AA contrast testing","bundle-budget discipline","hand-rolled SVG data visualization","print-fidelity export","mobile-first responsive redesign","schema hydration and backward-compatible persistence","open-source licensing and attribution"],"sourceFile":"proforma.use-case.md","id":"proforma#problem","heading":"Problem","excerpt":"AI business cases fall over in three predictable places. The benefit side gets estimated more generously than the cost side, because benefits are guesses and costs are invoices. The annual figures start in the wrong year, almost every fram…"},{"type":"case-study","project":"ProForma","slug":"proforma","route":"/case-studies/proforma","statusFamily":"production","tier":"primary","stack":["React 19","TypeScript","Vite","Vitest","hand-rolled SVG","localStorage","ESLint","Prettier","GitHub Actions"],"skills":["verified calculation engines","regression testing against a source of truth","CI-enforced WCAG AA contrast testing","bundle-budget discipline","hand-rolled SVG data visualization","print-fidelity export","mobile-first responsive redesign","schema hydration and backward-compatible persistence","open-source licensing and attribution"],"sourceFile":"proforma.use-case.md","id":"proforma#what-i-built","heading":"What I Built","excerpt":"A browser tool that models the five-year case and never sends the inputs anywhere. No account, no backend, no telemetry. Business cases carry salary figures and headcount plans, and the simplest way to be trustworthy with that data is to n…"},{"type":"case-study","project":"ProForma","slug":"proforma","route":"/case-studies/proforma","statusFamily":"production","tier":"primary","stack":["React 19","TypeScript","Vite","Vitest","hand-rolled SVG","localStorage","ESLint","Prettier","GitHub Actions"],"skills":["verified calculation engines","regression testing against a source of truth","CI-enforced WCAG AA contrast testing","bundle-budget discipline","hand-rolled SVG data visualization","print-fidelity export","mobile-first responsive redesign","schema hydration and backward-compatible persistence","open-source licensing and attribution"],"sourceFile":"proforma.use-case.md","id":"proforma#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"Verified calculation engines and regression testing against a source of truth, CI-enforced accessibility (WCAG AA contrast measured against the live stylesheet tokens rather than asserted), bundle-budget discipline and dependency minimalis…"},{"type":"case-study","project":"ProForma","slug":"proforma","route":"/case-studies/proforma","statusFamily":"production","tier":"primary","stack":["React 19","TypeScript","Vite","Vitest","hand-rolled SVG","localStorage","ESLint","Prettier","GitHub Actions"],"skills":["verified calculation engines","regression testing against a source of truth","CI-enforced WCAG AA contrast testing","bundle-budget discipline","hand-rolled SVG data visualization","print-fidelity export","mobile-first responsive redesign","schema hydration and backward-compatible persistence","open-source licensing and attribution"],"sourceFile":"proforma.use-case.md","id":"proforma#outcome-evidence","heading":"Outcome / Evidence","excerpt":"96 tests across 4 files, all measured rather than quoted: 41 on the calculation engine, 21 on colour contrast, 21 on formatting, 13 on document hydration. CI runs install, lint, format check, typecheck, test and build on every push and pul…"},{"type":"case-study","project":"ProForma","slug":"proforma","route":"/case-studies/proforma","statusFamily":"production","tier":"primary","stack":["React 19","TypeScript","Vite","Vitest","hand-rolled SVG","localStorage","ESLint","Prettier","GitHub Actions"],"skills":["verified calculation engines","regression testing against a source of truth","CI-enforced WCAG AA contrast testing","bundle-budget discipline","hand-rolled SVG data visualization","print-fidelity export","mobile-first responsive redesign","schema hydration and backward-compatible persistence","open-source licensing and attribution"],"sourceFile":"proforma.use-case.md","id":"proforma#interview-talking-points","heading":"Interview Talking Points","excerpt":"The regression suite is the point. Verifying arithmetic against a workbook once is a task; encoding those exact figures as tests so a later refactor cannot move them is engineering practice. The proof that it worked is that the engine touc…"},{"type":"case-study","project":"M3n0ko0g Website","slug":"m3n0ko0g-website","route":"/case-studies/m3n0ko0g-website","statusFamily":"production","tier":"primary","stack":["React 18","TypeScript 5","Vite 6","Tailwind CSS 3","React Router 7","Netlify Functions","OpenAI API (server-side)","Vitest"],"skills":["retrieval design without a vector database","BM25-style keyword and tag scoring","build-time corpus indexing","streaming LLM responses","graceful degradation to a deterministic search-only mode","serverless functions","secure server-side key handling","accessibility (reduced-motion, focus management, WCAG AA in two themes)","design-token discipline"],"sourceFile":"M3n0ko0g-Website.use-case.md","id":"m3n0ko0g-website#summary","heading":"M3n0ko0g Website (Agent Console + Digital Twin)","excerpt":"M3n0ko0g Website (Agent Console + Digital Twin). My portfolio rebuilt as an agent console, where a digital twin answers questions about my work by actually searching a build-time corpus in front of you, and the trace you watch is the retri…"},{"type":"case-study","project":"M3n0ko0g Website","slug":"m3n0ko0g-website","route":"/case-studies/m3n0ko0g-website","statusFamily":"production","tier":"primary","stack":["React 18","TypeScript 5","Vite 6","Tailwind CSS 3","React Router 7","Netlify Functions","OpenAI API (server-side)","Vitest"],"skills":["retrieval design without a vector database","BM25-style keyword and tag scoring","build-time corpus indexing","streaming LLM responses","graceful degradation to a deterministic search-only mode","serverless functions","secure server-side key handling","accessibility (reduced-motion, focus management, WCAG AA in two themes)","design-token discipline"],"sourceFile":"M3n0ko0g-Website.use-case.md","id":"m3n0ko0g-website#problem","heading":"Problem","excerpt":"A portfolio has to answer a question a static page cannot: the follow-up. A hiring manager reads \\"built an ETL pipeline\\" and wants to know whether it was idempotent, how big the feed was, and what happened when it failed. A founder wants t…"},{"type":"case-study","project":"M3n0ko0g Website","slug":"m3n0ko0g-website","route":"/case-studies/m3n0ko0g-website","statusFamily":"production","tier":"primary","stack":["React 18","TypeScript 5","Vite 6","Tailwind CSS 3","React Router 7","Netlify Functions","OpenAI API (server-side)","Vitest"],"skills":["retrieval design without a vector database","BM25-style keyword and tag scoring","build-time corpus indexing","streaming LLM responses","graceful degradation to a deterministic search-only mode","serverless functions","secure server-side key handling","accessibility (reduced-motion, focus management, WCAG AA in two themes)","design-token discipline"],"sourceFile":"M3n0ko0g-Website.use-case.md","id":"m3n0ko0g-website#what-i-built","heading":"What I Built","excerpt":"A React and TypeScript single-page app whose primary surface is a working agent console, backed by a Netlify function, over a corpus built at build time from the same Markdown files my coding agents read. A build-time indexer instead of a …"},{"type":"case-study","project":"M3n0ko0g Website","slug":"m3n0ko0g-website","route":"/case-studies/m3n0ko0g-website","statusFamily":"production","tier":"primary","stack":["React 18","TypeScript 5","Vite 6","Tailwind CSS 3","React Router 7","Netlify Functions","OpenAI API (server-side)","Vitest"],"skills":["retrieval design without a vector database","BM25-style keyword and tag scoring","build-time corpus indexing","streaming LLM responses","graceful degradation to a deterministic search-only mode","serverless functions","secure server-side key handling","accessibility (reduced-motion, focus management, WCAG AA in two themes)","design-token discipline"],"sourceFile":"M3n0ko0g-Website.use-case.md","id":"m3n0ko0g-website#skills-demonstrated","heading":"Skills Demonstrated","excerpt":"Retrieval design without a vector database, hand-rolled BM25 with field boosting and result diversification, build-time corpus indexing and artifact generation, isomorphic pure modules shared between serverless and browser, streaming NDJSO…"},{"type":"case-study","project":"M3n0ko0g Website","slug":"m3n0ko0g-website","route":"/case-studies/m3n0ko0g-website","statusFamily":"production","tier":"primary","stack":["React 18","TypeScript 5","Vite 6","Tailwind CSS 3","React Router 7","Netlify Functions","OpenAI API (server-side)","Vitest"],"skills":["retrieval design without a vector database","BM25-style keyword and tag scoring","build-time corpus indexing","streaming LLM responses","graceful degradation to a deterministic search-only mode","serverless functions","secure server-side key handling","accessibility (reduced-motion, focus management, WCAG AA in two themes)","design-token discipline"],"sourceFile":"M3n0ko0g-Website.use-case.md","id":"m3n0ko0g-website#outcome-evidence","heading":"Outcome / Evidence","excerpt":"Live on Netlify at ljefferson-menoko-site.netlify.app. Verifiable in the repository: a build-time indexer emitting a server corpus, a sanitized client corpus, per-case-study JSON and the sitemap; a 455-line streaming chat function whose ev…"},{"type":"case-study","project":"M3n0ko0g Website","slug":"m3n0ko0g-website","route":"/case-studies/m3n0ko0g-website","statusFamily":"production","tier":"primary","stack":["React 18","TypeScript 5","Vite 6","Tailwind CSS 3","React Router 7","Netlify Functions","OpenAI API (server-side)","Vitest"],"skills":["retrieval design without a vector database","BM25-style keyword and tag scoring","build-time corpus indexing","streaming LLM responses","graceful degradation to a deterministic search-only mode","serverless functions","secure server-side key handling","accessibility (reduced-motion, focus management, WCAG AA in two themes)","design-token discipline"],"sourceFile":"M3n0ko0g-Website.use-case.md","id":"m3n0ko0g-website#interview-talking-points","heading":"Interview Talking Points","excerpt":"The trace is the whole point and it is only worth anything because it is real. Search events stream before the model call because retrieval genuinely happens first, and the component renders the actual queries, chunk counts and matched hea…"},{"id":"profile#resume-overview","type":"profile","heading":"Resume, Lawrence Jefferson II — Overview","stack":[],"skills":[],"sourceFile":"resume.md","excerpt":"<!-- Agent memory | source: official resume PDF, July 2026 | public-safe --> Title: AI Engineer | AI Governance and Agent Harness Research | Creator of the Ag3nt24 Framework Location: Clarkston, WA · [email protected] Links:…"},{"id":"profile#resume-professional-summary","type":"profile","heading":"Resume, Lawrence Jefferson II — Professional Summary","stack":[],"skills":[],"sourceFile":"resume.md","excerpt":"AI engineer and AI governance researcher who designs and ships production-grade intelligent systems, not just prototypes, and studies how agents and their harnesses are governed and audited. Hands-on across the full ML lifecycle: LLM integ…"},{"id":"profile#resume-technical-skills","type":"profile","heading":"Resume, Lawrence Jefferson II — Technical Skills","stack":[],"skills":[],"sourceFile":"resume.md","excerpt":"AI/ML: LLMs, RAG, Agentic AI, Prompt Engineering, Fine-Tuning (QLoRA), Model Evaluation, LLMOps, Responsible AI, Safety Guardrails, Human-in-the-Loop, Deep Learning (in progress) Frameworks/Tools: LangChain, Hugging Face, Vector Databases,…"},{"id":"profile#resume-experience","type":"profile","heading":"Resume, Lawrence Jefferson II — Experience","stack":[],"skills":[],"sourceFile":"resume.md","excerpt":"CEO, CTO and lead architect, classHuman AI LLC. Clarkston WA (Jul 2026 – Present): Leads classHuman AI\'s AI governance and auditing research on agents and agent harnesses (questions, methods and measures first, sources cited, no results ye…"},{"id":"profile#resume-education-certifications-summary-see-certifications-md-for-full-detail","type":"profile","heading":"Resume, Lawrence Jefferson II — Education & Certifications (summary, see certifications.md for full detail)","stack":[],"skills":[],"sourceFile":"resume.md","excerpt":"Degree in progress: Bachelor of Science in Artificial Intelligence, American Military University (in progress, expected May 2028) Degree: A.S., Web Page, Digital/Multimedia & Information Resources De

1sign, American Military University AI/ML…"},{"id":"profile#profile-overview","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — Overview","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"<!-- Agent memory | public-safe: all content below is from Lawrence\'s own public LinkedIn profile --> Headline: AI Engineer | AI Governance and Agent Harness Research | classHuman AI LLC | WDVA Veteran-Owned Small Business | Protocol Droid…"},{"id":"profile#profile-summary-his-own-words-condensed","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — Summary (his own words, condensed)","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"My father was writing software in the 1960s. The lesson came through without ever being said: data integrity is the whole game. If you can\'t trust the source, nothing built on it is trustworthy either. That\'s still how I build. CEO, CTO an…"},{"id":"profile#profile-what-i-am-available-for","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — What I am available for","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"It depends on who\'s asking. Hiring an engineer: open to AI engineering and AI governance roles, including AI Engineer, Senior AI/ML Engineer, Senior Backend / AI Integration Engineer roles. Full-stack TypeScript, JavaScript and Python, wit…"},{"id":"profile#profile-research","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — Research","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"An agent harness is the code around a model that makes it an agent: the loop, tools, permissions, memory, gates, human sign-off and the record. My research is AI governance and auditing, specializing in agents and agent harnesses. I do it …"},{"id":"profile#profile-experience-timeline","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — Experience Timeline","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"classHuman AI LLC. CEO, CTO and lead architect (Jul 2026–present), Clarkston WA: I run an AI research organization for AI governance and auditing, specializing in agents and agent harnesses. We study how agents are governed: the harness ar…"},{"id":"profile#profile-military-service-detail","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — Military service detail","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"Combat deployments and operations other than war: Aug 1990–Jun 1991: Operation Desert Shield / Desert Storm / Saber-Sword Nov 2004–Oct 2005: OIF II Nov 2006–Apr 2007: Operation Jumpstart, supporting USMC and U.S. Customs and Border Patrol,…"},{"id":"profile#profile-education","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — Education","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"Bachelor of Science in Artificial Intelligence, American Military University (in progress, expected May 2028) American Military University, A.S., Web Page, Digital/Multimedia & Information Resources Design (Apr 2009–May 2012) Utah Valley U…"},{"id":"profile#profile-certifications-as-listed-on-linkedin","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — Certifications (as listed on LinkedIn)","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"AI Agents with Model Context Protocol Specialization Proficient AI Engineer Google AI Professional Certificate Complete Python Developer in 2023: Zero to Mastery Build an AI Agent See `certifications.md` for the full credential detail behi…"},{"id":"profile#profile-how-the-background-connects-to-the-work","type":"profile","heading":"Professional Profile, Lawrence Jefferson II (from LinkedIn, September 2026) — How the background connects to the work","stack":[],"skills":[],"sourceFile":"profile.md","excerpt":"The through-line from my father\'s software to Ag3nt24 is one principle at different scales: data integrity as the first question, a system of record you can actually trust, and a human holding final authority. It\'s why Ag3nt24\'s receipts g…"},{"id":"profile#certifications-overview","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — Overview","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"<!-- Agent memory | source: certification records + Credly, updated 2026-08-09 | public-safe Mirrors classhuman.org/credentials — keep the two in sync. Publishing rule (CLASSHUMAN.md): degrees/programs in progress are listed; in-progress C…"},{"id":"profile#certifications-education","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — Education","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"Bachelor of Science in Artificial Intelligence - American Military University (in progress, expected May 2028) A.S., Web Page, Digital/Multimedia & Information Resources De

1sign - American Military University (2009–2012), earned while servi…"},{"id":"profile#certifications-programs-bootcamps","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — Programs & Bootcamps","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"Full Stack JavaScript Software Engineer - Utah Valley University / V School: mastery-based program - JavaScript, React, Node, Express, MongoDB, auth, full-stack capstones Web Publishing & Development - JavaScript - Syracuse University Tech…"},{"id":"profile#certifications-professional-certificates-specializations","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — Professional Certificates & Specializations","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"Proficient AI Engineer - Full Program Completion - Ed Donner / proficientaiengineer.com (Aug 2026): AI Engineer Core Track (LLM Engineering, RAG, QLoRA, Agents) · Agentic Track (Agent Architectures, Frameworks & MCP) · MLOps Track (Vercel,…"},{"id":"profile#certifications-certifications-courses","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — Certifications & Courses","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"PCAP - Certified Associate in Python Programming - Python Institute (Syracuse / IVMF) PCEP - Certified Entry-Level Python Programmer - Python Institute (Dec 2022) ITIL 4® Foundation - PeopleCert / ACI Learning / LeaderQuest (Mar 2022) Comp…"},{"id":"profile#certifications-ibm-skillsbuild-badges","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — IBM SkillsBuild Badges","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"IBM SkillsBuild series: Foundations in Generative AI · Introduction to Large Language Models · Prompt Engineering: Shaping Better AI Responses · Introduction to Retrieval-Augmented Generation · Retrieval-Augmented Generation for Enhanced A…"},{"id":"profile#certifications-recognition","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — Recognition","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"Scrimba \\"Portfolio of the Week\\" - May 2026"},{"id":"profile#certifications-in-progress","type":"profile","heading":"Technology Certifications & Education - Lawrence Jefferson II — In Progress","stack":[],"skills":[],"sourceFile":"certifications.md","excerpt":"Applied AI Specialization - Vanderbilt University / Coursera (in progress) AI Engineering from Scratch - open-source course at aiengineeringfromscratch.com (in progress) Claude Cowork and Code master class - in progress Claude Certified Ar…"},{"id":"writing#medium-0","type":"writing","heading":"Cruising on the Coding Highway: A Joyride with JavaScript and Python","stack":[],"skills":[],"sourceFile":"blogPosts.json","link":"https://medium.com/@jefftkddan/cruising-on-the-coding-highway-a-joyride-with-javascript-and-python-d28a1a4ea109","excerpt":"Cruising on the Coding Highway: A Joyride with JavaScript and Python. Published essay on Medium by Lawrence Jefferson II."},{"id":"writing#medium-1","type":"writing","heading":"JavaScript Loops and the Map Method: A Fun Exploration","stack":[],"skills":[],"sourceFile":"blogPosts.json","link":"https://medium.com/@jefftkddan/javascript-loops-and-the-map-method-a-fun-exploration-cb672439e94d","excerpt":"JavaScript Loops and the Map Method: A Fun Exploration. Published essay on Medium by Lawrence Jefferson II."},{"id":"writing#medium-2","type":"writing","heading":"Menoko OG’s Magical Python Quest: Loops and List Comprehensions","stack":[],"skills":[],"sourceFile":"blogPosts.json","link":"https://medium.com/@jefftkddan/menoko-ogs-magical-python-quest-unleashing-coding-sorcery-with-loops-and-list-comprehe

1nsions-22c3e8a0ab99","excerpt":"Menoko OG’s Magical Python Quest: Loops and List Comprehensions. Published essay on Medium by Lawrence Jefferson II."},{"id":"writing#medium-3","type":"writing","heading":"Transitioning from Combat Engineering to Software Engineering","stack":[],"skills":[],"sourceFile":"blogPosts.json","link":"https://medium.com/@jefftkddan/transitioning-from-combat-engineering-to-software-engineering-a-path-to-healing-and-growth-1be494cd7d18","excerpt":"Transitioning from Combat Engineering to Software Engineering. Published essay on Medium by Lawrence Jefferson II."},{"id":"writing#medium-4","type":"writing","heading":"A Click is just a Click: Three Stages of Learning Development the Bruce Lee Way","stack":[],"skills":[],"sourceFile":"blogPosts.json","link":"https://medium.com/@jefftkddan/a-click-is-just-a-click-three-stages-of-learning-development-the-bruce-lee-way-1a374d9efab9","excerpt":"A Click is just a Click: Three Stages of Learning Development the Bruce Lee Way. Published essay on Medium by Lawrence Jefferson II."},{"id":"writing#medium-5","type":"writing","heading":"Immutable and Mutable Data Types in Python and JavaScript","stack":[],"skills":[],"sourceFile":"blogPosts.json","link":"https://medium.com/@jefftkddan/title-unveiling-the-magic-immutable-and-mutable-data-types-in-python-and-javascript-3811cc451b9b","excerpt":"Immutable and Mutable Data Types in Python and JavaScript. Published essay on Medium by Lawrence Jefferson II."},{"id":"writing#medium-6","type":"writing","heading":"The Human Touch: Don’t Be Scared of AI","stack":[],"skills":[],"sourceFile":"blogPosts.json","link":"https://medium.com/@jefftkddan/the-human-touch-unveiling-the-true-power-behind-technology-dont-be-scared-of-ai-ef6fe5fd88bc","excerpt":"The Human Touch: Don’t Be Scared of AI. Published essay on Medium by Lawrence Jefferson II."},{"id":"writing#essay-architecture-under-trust-pressure","type":"writing","slug":"architecture-under-trust-pressure","heading":"Architecture Under Trust Pressure","stack":[],"skills":[],"sourceFile":"localEssays.json","excerpt":"How to make architecture choices when correctness and auditability matter more than velocity theater. Audience: Founders in fintech and compliance-sensitive product spaces. Themes: Trust systems, Reliability, Decision discipline. Status: o…"},{"id":"writing#essay-fractional-cto-first-30-days","type":"writing","slug":"fractional-cto-first-30-days","heading":"Senior Backend Engineer: First 30 Days","stack":[],"skills":[],"sourceFile":"localEssays.json","excerpt":"A practical operating model for technical triage, roadmap truth, and system stabilization in month one. Audience: Startups without embedded senior technical leadership. Themes: CTO operations, Delivery confidence, Risk triage. Status: draf…"},{"id":"writing#essay-ai-guardrails-for-real-teams","type":"writing","slug":"ai-guardrails-for-real-teams","heading":"AI Guardrails for Real Teams","stack":[],"skills":[],"sourceFile":"localEssays.json","excerpt":"Where agentic systems create leverage, where they add risk, and how to design bounded automation with clear ownership. Audience: Product and engineering leaders evaluating AI workflows. Themes: AI strategy, Human-in-the-loop, Operational c…"},{"id":"learn#openai","type":"learn","heading":"Free AI courses — OpenAI","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 22 free OpenAI resources (OpenAI Academy), listed on the Learn page at /learn: Introduction to Prompt Engineering; Advanced Prompt Engineering; Introducing GPT-5.2; ChatGPT Projects; ChatGPT & Reasoning; ChatGPT Searc…"},{"id":"learn#anthropic","type":"learn","heading":"Free AI courses — Anthropic","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 17 free Anthropic resources (Anthropic Academy · Claude docs & tutorials), listed on the Learn page at /learn: Claude Code in Action; How to Use Claude for Everyday Work Tasks; AI Fluency: Framework & Foundations; Bui…"},{"id":"learn#google","type":"learn","heading":"Free AI courses — Google","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 17 free Google resources (Google Skills · Grow with Google · Coursera), listed on the Learn page at /learn: Prompt Engineering Whitepaper; Introduction to GenAI; Introduction to LLMs; Introduction to Responsible AI; A…"},{"id":"learn#microsoft","type":"learn","heading":"Free AI courses — Microsoft","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 10 free Microsoft resources (Microsoft Learn · Coursera), listed on the Learn page at /learn: Introduction to Generative AI and LLMs; AI for Beginners; Comparing Different LLMs; Using Generative AI Responsibly; Prompt…"},{"id":"learn#harvard","type":"learn","heading":"Free AI courses — Harvard","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 8 free Harvard resources (HKS GenAI course · CS50), listed on the Learn page at /learn: Introduction to Generative AI; The Basics of Generative AI; Prompt Engineering; Introduction to AI with Python (CS50); Beyond Cha…"},{"id":"learn#mit","type":"learn","heading":"Free AI courses — MIT","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 8 free MIT resources (OpenCourseWare · MIT Open Learning), listed on the Learn page at /learn: AI 101; Introduction to Deep Learning; Artificial Intelligence Course; Introduction to Machine Learning; How to AI (Almost…"},{"id":"learn#nvidia","type":"learn","heading":"Free AI courses — NVIDIA","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 9 free NVIDIA resources (NVIDIA DLI · NVIDIA Academy), listed on the Learn page at /learn: Generative AI Explained; Agentic AI Explained; Building A Brain in 10 Minutes; AI for All: From Basics to GenAI Practice; Gett…"},{"id":"learn#youtube","type":"learn","heading":"Free AI courses — YouTube","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 7 free YouTube resources (University lectures & talks), listed on the Learn page at /learn: MIT: Foundation Models & Generative AI; Stanford: Machine Learning & Building LLMs; Stanford: Machine Learning Full Course; W…"},{"id":"learn#freecodecamp","type":"learn","heading":"Free AI courses — freeCodeCamp","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 9 free freeCodeCamp resources (Certifications · News · YouTube), listed on the Learn page at /learn: freeCodeCamp Curriculum; Full Stack Developer Certification; Scientific Computing with Python; Machine Learning with…"},{"id":"learn#w3schools","type":"learn","heading":"Free AI courses — W3Schools","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 14 free W3Schools resources (Web, Python, SQL, Git, DSA tutorials), listed on the Learn page at /learn: W3Schools Home; HTML Tutorial; CSS Tutorial; JavaScript Tutorial; TypeScript Tutorial; Python Tutorial; Node.js T…"},{"id":"learn#mdn","type":"learn","heading":"Free AI courses — MDN","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 7 free MDN resources (Mozilla Developer Network), listed on the Learn page at /learn: Learn Web Development; MDN Curriculum; HTML Reference; CSS Reference; JavaScript Reference; JavaScript Guide;

1 HTTP Reference. These…"},{"id":"learn#python","type":"learn","heading":"Free AI courses — Python","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 12 free Python resources (Python Institute · Edube · Real Python · python.org), listed on the Learn page at /learn: Python Institute; Edube Interactive: Python Essentials 1 (free course); Edube Interactive: all OpenED…"},{"id":"learn#jsts","type":"learn","heading":"Free AI courses — JS / TS / Node","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 13 free JS / TS / Node resources (Official docs and free books), listed on the Learn page at /learn: The Modern JavaScript Tutorial; Eloquent JavaScript (4th edition); The TypeScript Handbook; TypeScript Playground; T…"},{"id":"learn#paths","type":"learn","heading":"Free AI courses — Learning paths","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 15 free Learning paths resources (Odin · Exercism · roadmap.sh · CS50), listed on the Learn page at /learn: The Odin Project; Exercism; Developer Roadmaps (roadmap.sh); AI Engineer Roadmap; Backend Developer Roadmap; …"},{"id":"learn#scrimba","type":"learn","heading":"Free AI courses — Scrimba","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 8 free Scrimba resources (Interactive screencasts (free tier)), listed on the Learn page at /learn: Scrimba (free tier); Learn JavaScript; Learn HTML and CSS; Learn React; Learn TypeScript; Intro to AI Engineering; Fr…"},{"id":"learn#mooc","type":"learn","heading":"Free AI courses — Coursera / Udemy","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 13 free Coursera / Udemy resources (Audit-free and free-tier courses), listed on the Learn page at /learn: Coursera: Free Courses; Supervised Machine Learning: Regression and Classification (audit free); Machine Learn…"},{"id":"learn#aws","type":"learn","heading":"Free AI courses — AWS","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 10 free AWS resources (Skill Builder · Bedrock · Strands), listed on the Learn page at /learn: AWS Skill Builder (free tier); AWS Cloud Practitioner Essentials; Learn About AI (AWS Training); AWS AI Practitioner Learn…"},{"id":"learn#ibm","type":"learn","heading":"Free AI courses — IBM","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 10 free IBM resources (SkillsBuild · watsonx · Coursera), listed on the Learn page at /learn: IBM SkillsBuild; IBM SkillsBuild Learning Catalog; IBM SkillsBuild Learner Portal; IBM AI Training: watsonx, ML and Generat…"},{"id":"learn#mlhubs","type":"learn","heading":"Free AI courses — DL.AI / HF / fast.ai / Kaggle","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 16 free DL.AI / HF / fast.ai / Kaggle resources (Hands-on ML and LLM courses), listed on the Learn page at /learn: DeepLearning.AI Courses (short courses free); DeepLearning.AI Learning Platform; Hugging Face Learn; H…"},{"id":"learn#labs","type":"learn","heading":"Free AI courses — Labs & frameworks","stack":[],"skills":[],"sourceFile":"free-ai-courses.md","route":"/learn","excerpt":"Lawrence studies from 22 free Labs & frameworks resources (MCP · LangChain · OpenAI · Gemini · Microsoft · Stanford), listed on the Learn page at /learn: Model Context Protocol: Getting Started; MCP for Beginners (Microsoft); LangChain Aca…"}]'),a={meta:e,projects:t,chunks:i};export{i as chunks,a as default,e as meta,t as projects};