All of my technical tutorials, musings and developer rants
Playing catch-up on AI?
Skip the research—here are the 4 AI tools I actually use to run my business
Between a full-time job at WorkOS and solo-parenting two toddlers with zero screen time, I published 14,790 words across 9 posts in two days. 47 images, 12 PRs. The secret: chatops with two AI agents I drive in compressed windows.
Practical lessons from running hands-on technical workshops at AI Engineer World Fair SF, AI Engineering London, and beyond. What works, what doesn't, and why most conference workshops suck.
My AI assistant Hermes writes MDX content, generates hero images, uploads to Bunny CDN, commits to git, opens pull requests, and merges — all from a Discord conversation.
An appreciation post for my colleague and friend Nick Nisi — TypeScript wizard, conference organizer, and the best workshop partner I could ask for.
I gave a keynote about walking in the woods with AI. Then I stopped needing the laptop entirely. Now I direct AWS infrastructure and content agents from my phone on the trail.
I run two AI agents — Claude Code for infrastructure, Hermes for content — both controlled from my phone. Here's the split-brain architecture and why separation of concerns matters for AI agents too.
Solving the 'silent files' problem: how I built an event-driven bridge between Claude Code on my Mac and the Hermes agent running on EC2, using SSM RunCommand, Tailscale, and HMAC-signed webhooks.
How I moved my personal AI assistant to AWS with OpenTofu, Tailscale, and Discord — self-hosting Hermes Agent in the margins of a normal workday.
I sat down with Jack Bridger on the Scaling Devtools podcast at AI Engineering London to talk about how WorkOS builds with AI coding agents, the patterns behind portable skills, and where developer tooling is heading.
Nick Nisi and I ran an 80-minute hands-on workshop at AI Engineering London on building Claude Code skills that are portable, executable, and composable. We taught constraints over instructions, evidence over guesses, and measurement over vibes.
I gave a talk at AI Engineering London about staying healthy, creative, and shipping while working with AI coding agents. The core idea: the agents scale infinitely, but your nervous system doesn't. Here's the full recap.
Claude Code's Remote Control feature lets you continue local terminal sessions from your phone. Here's how it works, what it can and can't do, and how it compares to Claude Code on the web.
A running tally.
I talk to my AI assistant like a person. I give it access to everything. And no, I don't think it's sentient. Here's why that's the point.
I built a watchOS app that lets me speak to any of my Claude Code sessions from my wrist. When you can't sit still, bring your AI with you.
Why I built a native Swift installer instead of writing another setup guide
A single developer can now direct multiple agentic coding agents across maintenance tasks while simultaneously building features in Cursor. Here's how the Orchestrator Pattern changes software development.
How I'm systematically teaching an AI system to compensate for specific ADHD/autism processing patterns—building collaborative intelligence that works with my neurology.
How I use Claude as an external brain and assistant—combining voice, agents, and a hardened CI/CD lane—to support an autistic mind with severe ADHD.
A DevSecCon 2025 keynote summary: orchestrate voice, agents, and a hardened CI/CD lane so you can think where you think best—while background agents safely ship production-grade code.
This may indeed be a bigger deal than MCP
Turn your health data into actionable insights with AI-powered analysis. Learn how to integrate your Oura Ring with Claude Desktop using the Model Context Protocol.
How LLMs mirrored my neurodivergent thinking, helped me connect the dots, and became the glue in a personal cognitive toolkit.
How ChatGPT and Google AI Studio helped me reduce live coding anxiety through systematic practice
A detailed breakdown of how I optimized my Next.js site's build process, moving from static generation to ISR, migrating images to CDN, and implementing on-demand OpenGraph image generation.
I got hands-on with Cursor's new Agents feature. Here's what I thought...
The corporate AI catch-22: move fast and break things, or move slow and get disrupted. Lessons from real-world AI enablement.
Nick Nisi and I taught 70 engineers how to build AI workflows and agents with Mastra.ai at the AI Engineer World Fair in San Francisco.
On May 20th, 2025, I gained access to Google Jules research preview. Here is what I think
On May 16th, 2025, I gained access to OpenAI's Codex research preview. Here is what I think
It is easier than ever to learn development, but the folks getting the most leverage are still seasoned engineers. Why?
$49💎 PremiumHow to code safely and efficiently with Git, secrets, and Cursor setup.
A no-jargon crash-course that keeps your AI side-projects from torching your wallet. Perfect for vibe coders with zero engineering background.
The Vercel AI SDK takes some time to get used it. It is time well spent.
Meet Gabbee.io, an AI BDR that programmatically validates leads via human-sounding phone calls. Automatically qualifies prospects using BANT methodology and syncs directly with Zoho CRM.
$149💎 PremiumStep by step tutorial: how to build a production-ready RAG pipeline.
$5💎 PremiumYou can automate your git profile switching based on the current directory. Here's how to do it.
Learn how I grew my tech blog to 35,000 readers in one year through site optimization, UX improvements, consistent publishing, and building useful tools.
I tried a handful of services when I last needed to fine-tune an LLM, and I was mostly disappointed...
A step by step tutorial with companion notebook.
One of the better Jupyter Notebooks to GPU-backed environment experiences I've had...
I've been on an MLOps adventure lately, taking any excuse to get hands on with neural nets, fine-tuning, Hugging Face datasets and models.
LoRA and QLoRA are two important innovations related to fine-tuning large language models like Llama and GPT.
AI assisted developer tooling is not created equally...
I have been experimenting with AI assisted dev tools nonstop. Cursor probably had the biggest impact of all, so far.
A comprehensive comparison and review of AI-assisted developer tools, including code autocompletion, intelligent terminals/shells, and video editing tools.
I trained a neural net to recognize hand-drawn digits, then built a Next.js UI for it
In the past 4 years, I have bought and used 3 different system76 linux machines. Here's why.
Embeddings models are the secret sauce that makes RAG work. How are THEY made?
For several years now I've been getting up early before work to get plenty of me time in...
You can chat with my writing and ask me questions I've already answered even when I'm not around
I wrote the RAG evaluation chapter for Pinecone's latest book
I wrote the CI/CD chapter for Pinecone's latest book
I want comfortable hearing aids with long battery life that extend my senses and intelligence
Why do I use an open-source IDE-like Neovim experience glued together with Lua?
It's been about a year since I last looked at Codeium - what has changed?
Sometimes someone will ask me how I am able to write new content so quickly. This is my answer.
A brief story about the best programmer I ever worked with
To effectively maintain richer and more complex web experiences, separate data from presentation.
The Warp terminal is a serious boon for anyone looking for help on the command line, but it's not yet perfect
The best thing about being a developer is being able to build your own tools, to your exact specifications.
Some junior engineers are so afraid of breaking something, they paradoxically stagnage themselves
A detailed comparison of the Pinecone and Chroma vector databases
A detailed comparison of the Pinecone and FAISS vector databases
A detailed comparison of the Pinecone and Milvus vector databases
A detailed comparison of the Pinecone and Weaviate vector databases
One of the best engineering managers I ever had was at Cloudflare. Here's what he did in his free time...
A detailed comparison of the Weaviate and Chroma vector databases
A detailed comparison of the Weaviate and FAISS vector databases
A detailed comparison of the Weaviate and Milvus vector databases
aws-vault is an excellent open-source tool written in Golang that allows you to manage the credentials for multiple AWS accounts securely.
I recently needed to figure out how to have my Vercel API route accept a new job and return a response immediately while still performing long running processing in the background. This is what I learned.
You can run git wherever JavaScript is accepted. Should you?
Dimensionality refers to the number of features a given embedding model extracts
Embeddings are numerical representations of the key features of some data
How do the different vector databases compare?
Steal my implementation to save yourself some time and headaches
What started out as an extended conversation with my mentor about career trajectory became a book!
What's the difference between Codeium and ChatGPT and which should you use?
A step-by-step walkthrough on how to generate arbitrary system load and flex Pinecone Serverless
I had a lot of growth and success to celebrate professionally in 2023
Developer delight === project velocity ?
Pageripper API uses Puppeteer and headless Chrome under the hood to see the same thing your browser does, even for Single Page Applications (SPAs)
I built an API that extracts data from webpages, even if they're rendered with Javascript. It's defined via Pulumi and deployed on AWS.
I introduced the new Pinecone AWS Reference Architecture with Pulumi and explained infrastructure as code
A comic strip about casting magic spells
A wordless comic strip about a typically brutal day at work, that nevertheless has a positive ending
Deploy production-ready systems using Pinecone in minutes with the AWS Reference Architecture
An examination of the Reference Architecture components and functionality
Design your side projects, blog posts and even your fun experiments to triangulate multiple learning paths simultaneously. Then, use them to build out your portfolio.
Control your own destiny, build your personal brand, and master web technologies by running your own tech blog.
There are only two people I am aware of named Zachary Proser, with that exact spelling. This article is only for Zachary Proser.
I have used StableDiffusion, AUTOMATIC111, DALLE and Discord bots to generate images in every style for blog posts. You can too.
Going the extra mile only to be unrewarded by your company feels like a personal slight and a waste of your time. It is not.
As of October 10th, 2023, ChatGPT4 and Codeium are still my favorite AI-assisted dev tool stack. Here is why
I have evaluated everything from ChatGPT to CoPilot to Codeium to Cursor, to Sourcegraph Cody to CodiumAI. All suffer from the same bugs
I got Copilot access for free as an active GitHub open-source maintainer, but would I pay for it?
These common mistakes are dooming your AI-assisted developer tooling startup to failure. Don't make them
Practices and protocols that keep me humming along, healthy and productive on a given day
CodiumAI's PR-agent integration leverages Pinecone under the hood to perform semantic search for similar GitHub issues
Programmers are people, and this is how I feel when...
If you're coming from Hacker News, I've learned that I need to spell out at the beginning that this post is partly facetious...
A history of package management and virtualenvs in Python and why we migrated the Pinecone Python client to Poetry
Jupyter Notebooks are surprisingly easy to get started with - especially when using GitHub and Google Colab
I built a custom opengraph image with '@vercel/og' that includes a fallback image for index pages
6 Technology trends position JavaScript developers to reap the rewards of the GenAI boom
How to reduce hallucinations in your Generative AI applications
Making the pivot from pure software engineering to developer advocacy
Stop yourself from committing a secret to git with ggshield
I get this question a lot: how do I get into coding? This is my best advice.
I open sourced my next.js AI chatbot and the Jupyter notebooks I used to build it, plus created a video series walking through it all on YouTube
Generally unimpressed with GitHub's Copilot (not Copilot X), I gave the alternative, Codeium a shot.
I have open sourced my automations project, which is a collection of shell scripts that automatically handle git operations, provide local code reviews, pull requests, and more!
Autogit is an open source shell script that you can wire up to your `cd` command to ensure you always have the latest code, branches, and remotes when you open up a local git repository.
If you're like me, you can't type your complex password correctly when your entire team is staring at you on a pair call. And now, you no longer have to.
After re-doing my personal website in next.js and deploying to Vercel, I reflect on how much better the developer experience has become.
Combining a lightweight state machine plus the Bubbletea charm library leads to a very powerful pattern for tooling that needs to orchestrate slow or expensive steps.
Join me as I delve into the infuriating, yet enlightening journey of maintaining my own Javascript-heavy website. Learn how battling DNS issues, dependency chaos, niche CSS pre-processors and constant painful upgrades has led to one of the most rewarding projects I've ever created.
As a Senior Software Engineer, I'm always looking for ways to refine my skills and optimize my workflow. This weekend, I experimented with integrating ChatGPT-4 into my developer toolkit alongside GitHub Copilot, which I've been using for several months. The goal? To see if these AI-powered tools could help me complete a side project more quickly.
In the past 3 years, I've been promoted 3 times. I reflect on the habits and activities that helped me improve the most.
A tutorial on how to write and use shell functions to fetch your tokens from the BitWarden CLI with one command.
Teatutor is a Golang CLI leveraging the Bubbletea Terminal User Interface (TUI) library from Charm.sh. It can be served over an SSH connection.
A full technical deep-dive on my optimizer app and its features
Pro tips for working with CircleCI configurations
As a developer, I often want to test my code on Windows, but don't have a Windows box handy. This tutorial demonstrates how to use Infrastructure as Code (IaC) to make provisioning Windows test instances easier. In this post, I also provide a working Packer template and Terraform configuration to deploy your own Windows test instance, as well as instructions for mounting a local folder over RDP.
I obssess a decent amount over my own developer productivity and my customized tmux, neovim and awesome window manager linux setup, and now I pass my best learnings on to you
A tutorial, along with working Terraform code and Dockerfile, that you can use to deploy and operate your own Quake 3 server in an ECS task.
Git-xargs allows you to run commands and scripts against many Github repos simultaneously
An open-source example repository and technical deep-dive on using AWS SAM, Golang, CodePipeline and CloudFormation to automate continuous delivery.
A ridiculously over-engineered CatFacts prank written in Golang and deployed via Kubernetes
I open sourced my first HTML5 game as a resource for other developers working with Phaser.js or wanting to build their own game
PICK UP THE PHONE - ITS CATFACTS!
A technical deep dive on building a portfolio site that is beautiful, blazing fast and 100% SEO optimized
A wikipedia crawling application