PageSourceSearch

https://ramisayar.com/component---src-templates-post-template-js-c…mis-readings-100-md-adf50b3464a5c135c79c.js

js ramisayar.com collected 2026-10-03 23:55:20 UTC 35,152 bytes, 2 lines download raw bytes

1"use strict";(self.webpackChunkramisayar_staticsite=self.webpackChunkramisayar_staticsite||[]).push([[4631],{230:function(e,t,a){a.r(t),a.d(t,{Head:function(){return p},default:function(){return g}});var n=a(9694),r=a(758),o=a(6998);function l(e){const t=Object.assign({p:"p",span:"span",a:"a",hr:"hr",h2:"h2",h4:"h4"},(0,n.RP)(),e.components);return r.createElement(r.Fragment,null,r.createElement(t.p,null,"Welcome to Rami’s Readings ",r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<code class="color-highlight" style="background-color: #100; color: #fff">#100</code>'}})," - a weekly digest of interesting articles, papers, videos, and X threads from my various sources across the Internet. Expect a list of reads covering AI, technology, business, culture, fashion, travel, and more. Learn about what I do at ",r.createElement(t.a,{href:"https://ramisayar.com/about?utm_source=newsletter&utm_medium=email",target:"_blank",rel:"nofollow noopener noreferrer"},"ramisayar.com/about"),"."),"\n",r.createElement(t.hr),"\n",r.createElement(o.A),"\n",r.createElement(t.hr),"\n",r.createElement(t.h2,{id:"-celebrating-100-newsletters-",style:{position:"relative"}},r.createElement(t.a,{href:"#-celebrating-100-newsletters-","aria-label":" celebrating 100 newsletters  permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"🎉 Celebrating 100 Newsletters 🎉"),"\n",r.createElement(t.p,null,"When I started this newsletter in January 2023, I had no idea where it would lead. I thought maybe a few curious minds would tune in. But here we are, 100 editions later with a community I couldn’t have imagined. ❤️"),"\n",r.createElement(t.p,null,"First, a massive THANK YOU! Whether you’ve been here since ",r.createElement(t.a,{href:"https://readings.ramisayar.com/p/ramis-readings-1",target:"_blank",rel:"nofollow noopener noreferrer"},"#1")," or just joined last week, your engagement, insights, and shares fuel this newsletter. Every reply and ",r.createElement(t.a,{href:"https://www.linkedin.com/posts/knutrisvik_rami-sayar-substack-activity-7290758183034195969-bj-R?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAFc-0IBIeShWjvXlGjwkHp7-W7abNDXAy8",target:"_blank",rel:"nofollow noopener noreferrer"},"testimonial")," ",r.createElement(t.a,{href:"https://www.linkedin.com/posts/activity-7289704243043700736-Nd2C?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAFc-0IBIeShWjvXlGjwkHp7-W7abNDXAy8",target:"_blank",rel:"nofollow noopener noreferrer"},"on")," ",r.createElement(t.a,{href:"https://www.linkedin.com/posts/portia-lane-child_ramis-readings-94-5-ai-predictions-activity-7296561944684036096-QjUX?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAFc-0IBIeShWjvXlGjwkHp7-W7abNDXAy8",target:"_blank",rel:"nofollow noopener noreferrer"},"LinkedIn")," pushes me to make this newsletter even better."),"\n",r.createElement(t.p,null,"Who reads this newsletter? It’s humbling to see leaders from Microsoft, Google, Amazon, S&P, and countless startups, VCs, and thought leaders tuning in every week. I am always surprised by who signs up. I’m sorry if you see me snooping on your LinkedIn after you subscribe, I’m just so curious to learn more about you - it is what inspired ",r.createElement(t.a,{href:"https://ramisayar.com/apero-and-intellect",target:"_blank",rel:"nofollow noopener noreferrer"},"Apéro & Intellect"),"."),"\n",r.createElement(t.p,null,"What started as a simple idea has turned into something way bigger. I never expected the reach, influence, and sheer momentum this would gain. If there’s one lesson I’ve learned, it’s this: consistency compounds."),"\n",r.createElement(t.h4,{id:"marking-100-with-a-special-apéro--intellect",style:{position:"relative"}}
1,r.createElement(t.a,{href:"#marking-100-with-a-special-ap%C3%A9ro--intellect","aria-label":"marking 100 with a special apéro  intellect permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Marking ",r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<code class="color-highlight" style="background-color: #100; color: #fff">#100</code>'}})," with a Special Apéro & Intellect"),"\n",r.createElement(t.p,null,"To mark ",r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<code class="color-highlight" style="background-color: #100; color: #fff">#100</code>'}}),", I’m planning an Apero & Intellect gathering this Thursday at 7pm —think drinks, deep discussions, and a toast to the next 100. If you’re in Seattle, let’s make it happen. Reply if you’re interested!"),"\n",r.createElement(t.h4,{id:"help-me-get-to-1000-subscribers-",style:{position:"relative"}},r.createElement(t.a,{href:"#help-me-get-to-1000-subscribers-","aria-label":"help me get to 1000 subscribers  permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Help Me Get to 1,000 Subscribers! 🚀"),"\n",r.createElement(t.p,null,"We’re at a major milestone, but I want to set my sights on the next big goal: growing this community to 1,000 engaged subscribers!"),"\n",r.createElement(t.p,null,"If you’ve found value in these insights, here’s how you can help:"),"\n",r.createElement(t.p,null,"✅ Share this newsletter with a friend, colleague, or your network—especially those who should be paying attention to AI and large language models."),"\n",r.createElement(t.p,null,"✅ Forward this email or use the referral option and win a prize:"),"\n",r.createElement(t.p,null,r.createElement(t.a,{href:"https://readings.ramisayar.com/leaderboard?&utm_source=post",target:"_blank",rel:"nofollow noopener noreferrer"},"Refer a friend")),"\n",r.createElement(t.p,null,"✅ Tell them why you read it! A personal recommendation goes a long way."),"\n",r.createElement(t.p,null,"I believe the best ideas spread through communities, not algorithms—and you, my readers, are the best ambassadors. Let’s grow this together."),"\n",r.createElement(t.p,null,"And as a thank you, I’m planning a special subscriber-only event once we hit 1,000. More details soon! 😉"),"\n",r.createElement(t.hr),"\n",r.createElement(t.h2,{id:"-10-ai-lessons-from-100-newsletters",style:{position:"relative"}},r.createElement(t.a,{href:"#-10-ai-lessons-from-100-newsletters","aria-label":" 10 ai lessons from 100 newsletters permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"🤖 10 AI Lessons From 100 Newsletters"),"\n",r.createElement(t.p,null,"After writing 100 issues of this newsletter, I’ve learned a lot about technology and how people are thinking about it. I am privileged to put much of what I’ve learned into ",r.createElement(t.a,{href:"https://ramisayar.com/about",target:"_blank",rel:"nofollow noopener noreferrer"},"practice"),". AI & LLMs aren’t just a passing trend;
1 they are a fundamental shift in how we work, create, and innovate. But with all the noise, the real challenge is separating hype from reality and figuring out what matters. To mark this milestone, I’m sharing 10 key lessons I’ve learned. Let’s dive in."),"\n",r.createElement(t.h4,{id:"lesson-1-developer-experience-is-driving-growth",style:{position:"relative"}},r.createElement(t.a,{href:"#lesson-1-developer-experience-is-driving-growth","aria-label":"lesson 1 developer experience is driving growth permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #1: Developer Experience Is Driving Growth"),"\n",r.createElement(t.p,null,"Across the entire AI stack, the models, tools, and libraries gaining the most mindshare prioritize the developer experience of Applied AI engineers. Anthropic open-sourced ",r.createElement(t.a,{href:"https://www.anthropic.com/news/model-context-protocol",target:"_blank",rel:"nofollow noopener noreferrer"},"MCP"),". ",r.createElement(t.a,{href:"https://www.langchain.com",target:"_blank",rel:"nofollow noopener noreferrer"},"LangChain")," and ",r.createElement(t.a,{href:"https://www.llamaindex.ai",target:"_blank",rel:"nofollow noopener noreferrer"},"LlamaIndex")," made it easy to get started with RAG. ",r.createElement(t.a,{href:"https://huggingface.co",target:"_blank",rel:"nofollow noopener noreferrer"},"HuggingFace"),"’s open source-first approach simplifies finding and distributing models. ",r.createElement(t.a,{href:"http://together.ai",target:"_blank",rel:"nofollow noopener noreferrer"},"Together.ai")," is growing rapidly with an emphasis on easy deployments. ",r.createElement(t.a,{href:"https://ollama.com",target:"_blank",rel:"nofollow noopener noreferrer"},"Ollama")," and ",r.createElement(t.a,{href:"http://lmstudio.ai",target:"_blank",rel:"nofollow noopener noreferrer"},"LMStudio")," are my go-to tools running models on my machines. Make it easy for developers to adopt your solution."),"\n",r.createElement(t.h4,{id:"lesson-2-prompt-engineering-is-superior-to-finetuning",style:{position:"relative"}},r.createElement(t.a,{href:"#lesson-2-prompt-engineering-is-superior-to-finetuning","aria-label":"lesson 2 prompt engineering is superior to finetuning permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #2: Prompt Engineering Is Superior to Finetuning"),"\n",r.createElement(t.p,null,"The reality is that prompt engineering enables much faster iteration than fine-tuning. While services like ",r.createElement(t.a,{href:"https://unsloth.ai",target:"_blank",rel:"nofollow noopener noreferrer"},"Unsloth")," are simplifying fine-tuning, prompt engineering still offers a superior developer experience (Lesson #1). It’s far easier to learn how to communicate effectively with existing models than to retrain them. My favorite resource for learning how to do effective prompt engineering is ",r.createElement(t.a,{href:"https://github.com/dair-ai",target:"_blank",rel:"nofollow noopener noreferrer"},"DAIR.AI"),"’s ",r.createElement(t.a,{href:"https://www.promptingguide.ai",target:"_blank",rel:"nofollow noopener noreferrer"},"Prompt Engineering Guide"),"."),"\n",r.createElement(t.h4,{id:"lesson-3-ai-is-getting-cheaper-faster-than-anticipated",style:{position:"relative"}}
1,r.createElement(t.a,{href:"#lesson-3-ai-is-getting-cheaper-faster-than-anticipated","aria-label":"lesson 3 ai is getting cheaper faster than anticipated permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #3: AI Is Getting Cheaper Faster Than Anticipated"),"\n",r.createElement(t.p,null,"I can’t believe how cheap it has become to run AI models, both locally and in the cloud. Two years ago, hosting and running models required massive CapEx investments in GPUs. Today, we’re running models that are significantly more powerful at a fraction of the previous cost per token."),"\n",r.createElement(t.p,null,"The cost reduction isn’t only being driven by competition among cloud and model providers, but also by fundamental engineering breakthroughs under the hood: ",r.createElement(t.a,{href:"https://arxiv.org/abs/2407.08608",target:"_blank",rel:"nofollow noopener noreferrer"},"Flash Attention"),", ",r.createElement(t.a,{href:"https://developer.nvidia.com/blog/introducing-new-kv-cache-reuse-optimizations-in-nvidia-tensorrt-llm/",target:"_blank",rel:"nofollow noopener noreferrer"},"KV Cache optimizations"),", and all the ",r.createElement(t.a,{href:"https://readings.ramisayar.com/i/158244576/deepseek-ai-released-deepgemm-smallpond-and-fs-engineering-fundamentals",target:"_blank",rel:"nofollow noopener noreferrer"},"magic")," DeepSeek did."),"\n",r.createElement(t.p,null,"Speaking of DeepSeek, I covered DeepSeek’s R1 ",r.createElement(t.a,{href:"https://readings.ramisayar.com/p/ramis-readings-95-welcome-and-more-on-deepseek",target:"_blank",rel:"nofollow noopener noreferrer"},"multiple")," ",r.createElement(t.a,{href:"https://readings.ramisayar.com/i/153558210/deepseek-r",target:"_blank",rel:"nofollow noopener noreferrer"},"times")," in this newsletter. While it shocked the broader stock market, the reality is that costs have been declining for years. In China, DeepSeekV2 triggered a staggering 90% price drop which I covered in ",r.createElement(t.a,{href:"https://readings.ramisayar.com/i/144974856/deepseekv-forcing-price-drops-to-of-previous-prices",target:"_blank",rel:"nofollow noopener noreferrer"},"#64"),"."),"\n",r.createElement(t.h4,{id:"lesson-4-users-arent-replacing-themselves-with-ai-theyre-amplifying-their-skills",style:{position:"relative"}},r.createElement(t.a,{href:"#lesson-4-users-arent-replacing-themselves-with-ai-theyre-amplifying-their-skills","aria-label":"lesson 4 users arent replacing themselves with ai theyre amplifying their skills permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #4: Users Aren’t Replacing Themselves With AI, They’re Amplifying Their Skills"),"\n",r.createElement(t.p,null,r.createElement(t.a,{href:"https://www.stlouisfed.org/on-the-economy/2025/feb/impact-generative-ai-work-productivity",target:"_blank",rel:"nofollow noopener noreferrer"},"Study")," after ",r.createElement(t.a,{href:"https://www.nber.org/papers/w31161",target:"_blank",rel:"nofollow noopener noreferrer"},"study")," shows that AI-driven productivity gains are benefiting workers at all skill levels—as long as they integrate AI into their workflows. Don’t buy into the hype that AI will replace you. The real shift is that AI amplifies productivity for those who know how to use it effectively. Form your own opinion and experience with each of the models and interfaces."),"\n",r.createElement(t.h4,{id:"lesson-5-knowing-when-to-automate",style:{position:"relative"}}
1,r.createElement(t.a,{href:"#lesson-5-knowing-when-to-automate","aria-label":"lesson 5 knowing when to automate permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #5: Knowing When to Automate"),"\n",r.createElement(t.p,null,"Lesson #4 naturally leads to Lesson #5: Knowing When to Automate. In a talk I gave at ",r.createElement(t.a,{href:"https://www.linkedin.com/posts/ramisayar_genai-mit-microsoftai-activity-7190737522233090049-bf1-?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAFc-0IBIeShWjvXlGjwkHp7-W7abNDXAy8",target:"_blank",rel:"nofollow noopener noreferrer"},"MIT"),", I broke down the key variables of LLM and AI systems—variables like cost (yes, even after the DeepSeek-driven price drops), quality of data and output, speed of the system, and delivery time all play a role in deciding whether automation makes sense. The reality is that every AI system requires tradeoffs between these variables. Not everything should be automated. The key is to evaluate whether the benefits outweigh the costs for your specific problem."),"\n",r.createElement(t.h4,{id:"lesson-6-own-your-data-and-benchmarks",style:{position:"relative"}},r.createElement(t.a,{href:"#lesson-6-own-your-data-and-benchmarks","aria-label":"lesson 6 own your data and benchmarks permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #6: Own Your Data and Benchmarks"),"\n",r.createElement(t.p,null,"“Data is the new oil” has been a popular analogy for years, but in today’s AI and LLM world: high-quality, structured, and proprietary data is the real competitive advantage. Raw data alone isn’t enough. Data requires constant cleaning, restructuring, re-engineering, and maintenance to retain its value. AI models are only as good as the data they are trained on, but owning your data also allows you to create your own benchmarks to evaluate new models that I constantly publicize in this newsletter (Sorry for making you work so often 😂). ",r.createElement(t.a,{href:"https://huggingface.co/collections/open-llm-leaderboard/the-big-benchmarks-collection-64faca6335a7fc7d4ffe974a",target:"_blank",rel:"nofollow noopener noreferrer"},"Public benchmarks")," are useful, but custom benchmarks tailored to your use cases are critical."),"\n",r.createElement(t.h4,{id:"lesson-7-companies-that-ignore-ai-will-be-outpaced",style:{position:"relative"}},r.createElement(t.a,{href:"#lesson-7-companies-that-ignore-ai-will-be-outpaced","aria-label":"lesson 7 companies that ignore ai will be outpaced permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #7: Companies That Ignore AI Will Be Outpaced"),"\n",r.createElement(t.p,null,"AI-native startups are achieving revenue-to-headcount ratios that have never been seen before. There’s even a ",r.createElement(t.a,{href:"https://leanaileaderboard.com",target:"_blank",rel:"nofollow noopener noreferrer"},"leaderboard")," tracking these companies.\nStartups are scaling using internal revenue, reducing their dependence on VC fundraising, and are no longer throttled by money cycles. This shift means they can move faster, stay lean, and outmaneuver slower incumbents weighed down by debt. Companies that ignore AI won’t just fall behind, they’ll be outpaced entirely by nimble, AI-powered competitors. A ",r.createElement(t.a,{href:"https://ocw.mit.edu/courses/15-871-introduction-to-system-dynamics-fall-2013/",target:"_blank",rel:"nofollow noopener noreferrer"},"System Dynamics")," analysis of your organization is what you need to make the most of AI."),"\n",r.createElement(t.h4,{id:"lesson-8-engineering-fundamentals-rule",style:{position:"relative"}}
1,r.createElement(t.a,{href:"#lesson-8-engineering-fundamentals-rule","aria-label":"lesson 8 engineering fundamentals rule permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #8: Engineering Fundamentals Rule"),"\n",r.createElement(t.p,null,"My first programming language was C++, and for half my career, it felt like maintaining that knowledge was a waste of time as Python and JavaScript dominated most projects. But today, I am seeing a clear pivot back to low-level, high-performance languages like C++ and Rust. Projects like ",r.createElement(t.a,{href:"https://github.com/ggml-org/llama.cpp",target:"_blank",rel:"nofollow noopener noreferrer"},"llama.cpp")," & ",r.createElement(t.a,{href:"https://github.com/ml-explore/mlx",target:"_blank",rel:"nofollow noopener noreferrer"},"MLX"),", and frameworks like CUDA, are making low-level languages a priority for developers. Rust is even replacing JavaScript in JavaScript build tools (see ",r.createElement(t.a,{href:"https://swc.rs",target:"_blank",rel:"nofollow noopener noreferrer"}
1,"SWC"),")! The shift is clear. The software engineering fundamentals of performance, memory efficiency, and parallelism matter more than ever in modern development."),"\n",r.createElement(t.h4,{id:"lesson-9-edge-ai-is-the-future",style:{position:"relative"}},r.createElement(t.a,{href:"#lesson-9-edge-ai-is-the-future","aria-label":"lesson 9 edge ai is the future permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #9: Edge AI Is The Future"),"\n",r.createElement(t.p,null,"As I shared in ",r.createElement(t.a,{href:"https://readings.ramisayar.com/p/ramis-readings-94-5-ai-predictions-for-2025",target:"_blank",rel:"nofollow noopener noreferrer"},"#95"),", open source LLMs optimized to run on devices with consumer-grade hardware are becoming the norm. ",r.createElement(t.a,{href:"https://ollama.com",target:"_blank",rel:"nofollow noopener noreferrer"},"Ollama")," and ",r.createElement(t.a,{href:"http://lmstudio.ai",target:"_blank",rel:"nofollow noopener noreferrer"},"LMStudio")," are making it so easy for any developer to just build local AI apps. At CES 2025, Edge AI was all the buzz with nearly every OEM marketing local AI-powered hardware. Apple’s M4 chip run state-of-the-art LLMs on battery power. In 2025, Edge AI is the future."),"\n",r.createElement(t.h4,{id:"lesson-10-dont-stop-reading-and-learning",style:{position:"relative"}},r.createElement(t.a,{href:"#lesson-10-dont-stop-reading-and-learning","aria-label":"lesson 10 dont stop reading and learning permalink",className:"anchor before"},r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<svg aria-hidden="true" focusable="false" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg>'}})),"Lesson #10: Don’t Stop Reading and Learning"),"\n",r.createElement(t.p,null,"Business environments and AI are evolving at a frenetic pace, and the engineers, founders, and builders who stay ahead are the ones who keep adapting. The fundamentals—data structures, algorithms, and systems engineering—still rule, but the tools and best practices are shifting fast. Stay curious. Experiment. Ship. Iterate. As I shared in ",r.createElement(t.a,{href:"https://readings.ramisayar.com/i/158244576/tyler-cowen-the-man-who-wants-to-know-everything",target:"_blank",rel:"nofollow noopener noreferrer"},"#99"),", I will do my best to keep reading and learning, and bring you, my beloved community, along for the ride!"),"\n",r.createElement(t.p,null,"Andiamo!"),"\n",r.createElement(t.hr),"\n",r.createElement(t.p,null,"Celebrate with me today by grabbing a cupcake!"),"\n",r.createElement(t.span,{dangerouslySetInnerHTML:{__html:'<span\n      class="gatsby-resp-image-wrapper"\n      style="position: relative; display: block; margin-left: auto; margin-right: auto; max-width: 640px; "\n    >\n      <a\n    class="gatsby-resp-image-link"\n    href="/static/2f8ea7be0efa398117773015170f0828/47311/ramis-readings-100-0.jpg"\n    style="display: block"\n    target="_blank"\n    rel="noopener"\n  >\n    <span\n    class="gatsby-resp-image-background-image"\n    style="padding-bottom: 66.875%; position: relative; bottom: 0; left: 0; background-image: url(\'data:image/jpeg;base64,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\'); background-size: cover; display: block;"\n  ></span>\n  <picture>\n          <source\n              srcset="/static/2f8ea7be0efa398117773015170f0828/81b0c/ramis-readings-100-0.avif 160w,\n/static/2f8ea7be0efa398117773015170f0828/297e5/ramis-readings-100-0.avif 320w,\n/static/2f8ea7be0efa398117773015170f0828/e6133/ramis-readings-100-0.avif 640w,\n/static/2f8ea7be0efa398117773015170f0828/542bc/ramis-readings-100-0.avif 960w,\n/static/2f8ea7be0efa398117773015170f0828/d8ca6/ramis-readings-100-0.avif 1080w"\n              sizes="(max-width: 640px) 100v
1w, 640px"\n              type="image/avif"\n            /><source\n              srcset="/static/2f8ea7be0efa398117773015170f0828/60a22/ramis-readings-100-0.webp 160w,\n/static/2f8ea7be0efa398117773015170f0828/4c812/ramis-readings-100-0.webp 320w,\n/static/2f8ea7be0efa398117773015170f0828/0ba47/ramis-readings-100-0.webp 640w,\n/static/2f8ea7be0efa398117773015170f0828/e46b2/ramis-readings-100-0.webp 960w,\n/static/2f8ea7be0efa398117773015170f0828/260c2/ramis-readings-100-0.webp 1080w"\n              sizes="(max-width: 640px) 100vw, 640px"\n              type="image/webp"\n            />\n          <source\n            srcset="/static/2f8ea7be0efa398117773015170f0828/0913d/ramis-readings-100-0.jpg 160w,\n/static/2f8ea7be0efa398117773015170f0828/cb69c/ramis-readings-100-0.jpg 320w,\n/static/2f8ea7be0efa398117773015170f0828/c08c5/ramis-readings-100-0.jpg 640w,\n/static/2f8ea7be0efa398117773015170f0828/6a068/ramis-readings-100-0.jpg 960w,\n/static/2f8ea7be0efa398117773015170f0828/47311/ramis-readings-100-0.jpg 1080w"\n            sizes="(max-width: 640px) 100vw, 640px"\n            type="image/jpeg"\n          />\n          <img\n            class="gatsby-resp-image-image"\n            src="/static/2f8ea7be0efa398117773015170f0828/c08c5/ramis-readings-100-0.jpg"\n            alt="ramis readings 100 0"\n            title=""\n            loading="lazy"\n            decoding="async"\n            style="width:100%;height:100%;margin:0;vertical-align:middle;position:absolute;top:0;left:0;"\n          />\n        </picture>\n  </a>\n    </span>'}}),"\n",r.createElement(t.p,null,"Signing off from Redmond, WA."))}var s=function(e){void 0===e&&(e={});const t=Object.assign({},(0,n.RP)(),e.components).wrapper;return t?r.createElement(t,e,r.createElement(l,e)):l(e)},i=a(7854),c=a(3207),h=a(3623),m=a(2305);const d=e=>{let t=e.data.mdx,a=e.children;return r.createElement(i.A,null,r.createElement(c.A,{post:t},a))},p=e=>{let t=e.location,a=e.data;if(!a.mdx)return null;const n=(0,m.Q6)(),o=n.title,l=n.subtitle,s=a.mdx.frontmatter,i=s.title,c=s.description,d=null!==c?c:l,p=i+" - "+o;return r.createElement(h.A,{title:p,description:d,pathname:t.pathname})};function g(e){return r.createElement(d,e,r.createElement(s,e))}},3207:function(e,t,a){a.d(t,{A:function(){return p}});var n=a(758),r=a(5662),o=a(8173),l=a(2305);var s=()=>{const e=(0,l.Q6)().author;return n.createElement("div",{className:"Author-module--author--1c58d"},n.createElement("p",{dangerouslySetInnerHTML:{__html:e.bio}}),n.createElement("p",null,n.createElement("a",{className:"Author-module--author__bio-twitter--75212",href:(0,o.e)("twitter",e.contacts.twitter),rel:"noopener noreferrer",target:"_blank"},n.createElement("strong",null,e.name)," on Twitter")))};var i=e=>{let t=e.title,a=e.children;return n.createElement("div",{className:"Content-module--content--80d58"},n.createElement("h1",{className:"Content-module--content__title--f2408"},t),n.createElement("div",{className:"Content-module--content__body--0ce51"},a))},c=a(6343),h=a(6219);var m=e=>{let t=e.date;return n.createElement("div",null,n.createElement("p",{className:"Meta-module--meta__date--0c1b6"},"Published ",(0,c.GP)((0,h.H)(t),"d MMM yyyy")))};var d=e=>{let t=e.tags,a=e.tagSlugs;return n.createElement("div",{className:"Tags-module--tags--18589"},n.createElement("ul",{className:"Tags-module--tags__list--7df03"},a&&a.map((e,a)=>n.createElement("li",{className:"Tags-module--tags__list-item--a981e",key:t[a]},n.createElement(r.Link,{to:e,className:"Tags-module--tags__list-item-link--3ec2a"},t[a])))))};var p=e=>{let t=e.post,a=e.children;if(!t)return null;const o=t.fields.tagSlugs,l=t.frontmatter,c=l.tags,h=l.title,p=l.date;return n.createElement("div",null,n.createElement(r.Link,{className:"Post-module--post__home-button--c395c",to:"/"},"All Articles"),n.createElement("div",null,n.createElement(i,{title:h},a)),n.createElement("div",{className:"Post-module--post__footer--0bca4"},n.createElement(m,{date:p}),c&&o&&n.createElement(d,{tags:c,tagSlugs:o}),n.createElement(s,null)))}},6998:function(e,t,a){a.d(t,{A:function(){return r}});var n=a(758);var r=()=>n.createElement("div",{className:"SubstackCTA-module--substack_cta--b6e61"},n.createElement("h2",{className:"SubstackCTA-module--substack_cta__header--13588"},"Want a peak inside my brain?"),n.createElement("p",{className:"SubstackCTA-module--substack_cta__text--bbc91"},"I write a newsletter called"," ",n.createElement("a",{href:"https://readings.ramisayar.com?utm_source=ramisayar.com&utm_medium=referral"},"Rami's Readings"),".",n.createElement("br",null),n.createElement("br",null),"Subscribe for a curated digest of exciting reads on LLMs, AI, economics, business, and tech, plus my brief expert notes. Trusted by leaders at Microsoft, Amazon, Google, Stripe, and MIT. Less noise, more hidden gems."),n.createElement("iframe",{className:"SubstackCTA-module--substack_cta__iframe--95253",src:"https://readings.ramisayar.com/embed",width:"325",height:"160",frameBorder:"0",scrolling:"no",title:"Substack"}))},9694:function(e,t,a){a.d(t,{RP:function(){return o}});var n=a(758);const r=n.createContext({});function o(e){const t=n.useContext(r);return n.useMemo(()=>"function"==typeof e?e(t):{...t,...e},[t,e])}}}]);
2//# sourceMappingURL=component---src-templates-post-template-js-content-file-path-content-posts-newsletter-ramis-readings-100-md-adf50b3464a5c135c79c.js.map

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.