From Pixels to Particles

In this blog post, I explore a collection of thought-provoking articles and topics. From the emerging role of Context Engineering in AI and the realistic future of software engineering with large language models to a deep dive into why the AI revolution will be slower than anticipated, there’s a lot to ponder. Plus, insights into the significance of curation as a new form of intelligence and why even powerful tools might not hold all the answers. Dive into these intriguing reads with me!

• The New Skill in AI is Not Prompting, It’s Context Engineering: Context Engineering is the new skill in AI. It is about providing the right information and tools, in the right format, at the right time.
• Software engineering with LLMs in 2025: reality check: How are devs at AI startups and in Big Tech using AI tools, and what do they think of them? A broad overview of the state of play in tooling, with Anthropic, Google, Amazon, and others
• The AI Revolution Won’t Happen Overnight: AI will transform industries. However, this transformation will happen on enterprise time: longer, slower, and with far more friction than most expect, and much slower than Silicon Valley is selling. Right now, companies are getting six fundamental things wrong about how AI will create value and how long it will take—and risk wasting resources, overpromising results, and eroding trust. 1) AI’s real impact will take much longer than we think. 2) We’re being wildly optimistic about enterprise AI adoption. 3) The market is overestimating the value of AI companies. 4) The real money isn’t in the models. 5) We’re over indexing on startups. 6) We’re obsessed with generative AI but it’s not the future. Enterprise leaders need to be asking fewer questions about what AI might do—and more about what it’s actually doing in your business.
• Taste Is the New Intelligence: Why curation, discernment, and restraint matter more than ever
• Toolmen: Even the best weapon is an unhappy tool.
• There Are No New Ideas in AI… Only New Datasets: LLMs were invented in four major developments… all of which were datasets

Hey there! In my latest post, I dive into some exciting tech projects I’ve been exploring. There’s the zen-mcp-server, an innovative blend of Claude Code, OpenAI, and more. I also stumbled upon octelium, a versatile platform acting as a VPN, secure access platform, and much more. Plus, I’ve been learning about Bloom Filters with a great tutorial. Check them out if you’re into tech like me!

• zen-mcp-server: The power of Claude Code + [Gemini / OpenAI / Grok / OpenRouter / Ollama / Custom Model / All Of The Above] working as one. – BeehiveInnovations/
• consult7:
• octelium: A next-gen FOSS self-hosted unified zero trust secure access platform that can operate as a remote access VPN, a ZTNA/BeyondCorp architecture, API/AI gateway, a PaaS, an infrastructure for MCP & A2A architectures or even as an ngrok-alternative and a homelab infrastructure. – octelium/octelium
• Bloom Filters by Example:

In my latest blog post, I’m highlighting an interesting project called xslt, which is a native web build system using XML and XSLT. If you’re curious about contributing to this project or just want to learn more, feel free to check it out on GitHub!

• xslt: Native web build system (XML+XSLT). Contribute to pacocoursey/xslt development by creating an account on GitHub.

In this blog post, I explore fascinating mathematical discoveries, ponder problem-solving wisdom from Richard Feynman, and delve into the intricacies of augmented coding into a technically challenging project. Join me as I uncover a new tetrahedron shape that defies gravity by always landing the same side up, and discover insights on approaching problems through Feynman’s lens, along with reflections on cutting-edge coding techniques.

• A New Pyramid-Like Shape Always Lands the Same Side Up: A tetrahedron is the simplest Platonic solid. Mathematicians have now made one that’s stable only on one side, confirming a decades-old conjecture.
• What Problems to Solve – By Richard Feynman:
• Augmented Coding: Beyond the Vibes: Notes from a technically challenging project

Hey there! In my latest blog post, I dive into some intriguing reads: Birgitta Böckeler shares insights from using AI in software engineering, and there’s a look into why engineers sometimes clash with managers. I explore the perks of vibe coding and why treating infrastructure as code is essential. Plus, I touch on reinforcement learning and how AI like ChatGPT is reshaping our language. Happy reading!

• Learnings from two years of using AI tools for software engineering: How to think about today’s AI tools, approaches that work well, and concerns about using them for development. Guest post by Birgitta Böckeler, Distinguished Engineer at Thoughtworks
• Why Engineers Hate Their Managers (And What to Do About It): Discover why engineers hate managers, the common management anti-patterns that destroy trust, and practical solutions from someone who’s been on both sides.
• Vibe coding: I see AI as a very valuable tool, but only that. Something that CAN help you if used correctly and in the right place, and can also cause issues when letting it run rampant. You can be both excited about a technology and careful with putting it to use in a way that brings actual value.
• Infrastructure as Code is a MUST have: Infrastructure as Code (IaC) certainly has become a buzzword so this serves as a quick guide or reminder of why IaC is so important. IaC is what the name says, defining your infrastructure as code…
• Reinforcement learning, explained with a minimum of math and jargon: To create reliable agents, AI companies had to go beyond predicting the next token.
• You sound like ChatGPT: AI isn’t just impacting how we write — it’s changing how we speak and interact with others. And there’s only more to come.
• Writing Toy Software Is A Joy: Why you should write more toy programs: “What I cannot create, I do not understand”

In my latest blog post, I dive into why I chose the ‘Interdependence Retirement Plan’ over a dream job, prioritizing community care instead of traditional wealth accumulation. Plus, I share insights from “The Magic of Code” by Samuel Arbesman, which explores the profound impact of digital language on our world and its potential to shape our future. Join me in rethinking retirement and technology!

• Interdependence is My New Retirement Plan: Why I turned down my dream job to pursue the ‘Interdependence Retirement Plan’ – choosing community care over conventional financial security and wealth accumulation.
• The Magic of Code: How Digital Language Created and Connects Our World–And Shapes Our Future by Samuel Arbesman: In the tradition of classics such as The Lives of a Cell, a bold reframing of our relationship wi…

Hey there! In my latest blog post, I dive into some intriguing topics. I start with a step-by-step guide to creating a personal website using AI in just 15 minutes without any coding! Also, I explore the world of AI system development and share insights on “vibe coding.” Plus, I’ve included a peek into a modern fitness coaching platform and tips on writing killer AI documentation. Check it out!

• Full Tutorial: Build a Stunning Personal Website with AI in 15 Minutes: From zero to published website using Bolt (no coding experience required)
• AI system development: LLM → RAG → AI Workflow → AI Agent: Confused by LLMs, RAG, & AI Agents? We break down the spectrum of AI system design with a familiar resume-screening example to show the leap in capabilities (and complexity) at each level.
• How to Vibe Code as a Senior Engineer: Over the past few months, I’ve rediscovered my love for building software. Not in the traditional way—writing line after line of code, plumbing together brittle libraries, grinding through boilerplate. Instead, I’ve been vibe coding. Vibe coding is a new way of building software where you use one
• Snouzy/workout-cool: 🏋 Modern open-source fitness coaching platform. Create workout plans, track progress, and access a comprehensive exercise database. –
• Writing documentation for AI: best practices: Retrieval-Augmented Generation (RAG) systems like Kapa rely on your
• My (mostly) minimalistic AI setup as a Senior Engineer in Big Tech: You can stop being overwhelmed by the hundreds of new tools every day

In my latest post, I’ve shared some intriguing topics that captured my attention. You’ll find a gem about AI pair programming right in your terminal, a fascinating account of resurrecting an old torrent tracker (yes, it’s Linux ISOs!), and an insightful look into Anthropic’s approach to developing reliable AI agents. Curious? Dive into the details and expand your digital horizons!

• AI pair programming in your terminal:
• Resurrecting a dead torrent tracker and finding 3 million peers: So I was uh, downloading some linux isos, like usual. It was going slowly, so I opened up the Trackers tab in qBittorrent and saw the following:
• Agents: Discover how Anthropic approaches the development of reliable AI agents. Learn about our research on agent capabilities, safety considerations, and technical framework for building trustworthy AI.

In this blog post, I share my thoughts on two interesting reads. The first one is Miguel Grinberg’s take on why generative AI coding tools and agents haven’t quite worked out for him. It’s a thought-provoking piece available on miguelgrinberg.com. The second is a deep dive on AI with Claude Code, featured on Kade Killary’s blog. Both are insightful perspectives worth exploring!

• Why Generative AI Coding Tools and Agents Do Not Work For Me: miguelgrinberg.com
• blog – kade@localhost:~$: blog – kade killary