Every Agent Session Should Be a Heist
I stopped asking one model to play every role. Pokémon Red showed me how to cast a crew by benchmark and route each job with vLLM.
I stopped asking one model to play every role. Pokémon Red showed me how to cast a crew by benchmark and route each job with vLLM.
My OpenClaw chief-of-staff agent was burning $80 a day in Anthropic API costs. The fix was one config line, and the only reason I found it was observability.
Skills are a great way to teach an agent a pattern. I got curious whether the agent could just learn the pattern, so I tried training a model while also figuring out what RL actually is.
I built a tool that fans out parallel Claude Code agents to fix lint errors across a codebase. It taught me that the hard problems aren't prompts or models. They're isolation, observability, and memory.
How a DeepMind paper turned my Pokemon agent from 'watch and tweak' into 'run and measure.'
I was an hour deep into a Postgres integration when Claude Code crashed. No session recovery. But tapes had every message stored in a local SQLite database.
Agent sessions contain valuable wisdom worth preserving. Like magnetic tape archives knowledge for decades, Tapes preserves agent memory for lasting value.
The Ralph Loop is having a moment. Developers are shipping code while they sleep. But there's a problem we're not talking about: nobody knows what Ralph learned along the way.
How moving to edge functions improved performance without a full rewrite.
How Continue's extensible platform goes beyond code generation to enable custom AI agents for code review at scale.
AI code reviews that run in your GitHub Actions environment, keeping your code in your repository. Built on Continue for flexible, privacy-focused code analysis.
How pre-commit hooks help AI assistants ship better code by focusing on implementation instead of formatting debates
A comprehensive guide to implementing workspace dashboard UI components using Storybook-first development methodology with real-world examples and technical decisions.
Learn how I reduced the main JavaScript bundle from 1,158KB to 204KB through advanced chunking strategies, improving LCP from 5.9 seconds while maintaining full functionality.
Discover how implementing a Supabase-based avatar caching strategy dramatically improved our Pull Request Contributions chart performance by reducing load times and eliminating redundant GitHub API calls.
Learn how I integrated Bluesky's social features into my blog, creating a zero-maintenance comment system using the AT Protocol's open APIs. This technical walkthrough covers the implementation details, UI design decisions, and lessons learned from building social engagement without databases or authentication.
Explore how implementing blur-up progressive image loading for avatar images significantly improves perceived performance and user experience, with technical implementation details and code examples.
Learn how to build reliable, fast, and maintainable tests for UI components without relying on mocks, improving test suite reliability and developer experience.
A two-model workflow using Claude Opus 4.1 for deep planning and Sonnet for rapid implementation.
Using Chroma Cloud to find semantically similar GitHub issues based on meaning, not just keyword matching.
Using Cognee MCP as an AI memory engine that learns from every interaction and remembers what you teach it.
Technical guide covering the implementation of an AI-powered prompt editor with text selection detection, sidebar interface, Supabase edge functions, and streaming OpenAI responses. Includes code examples, security considerations, and cost optimization strategies using GPT-4o tier models.
Using Recurse ML rules as automated quality gates for async AI development during fragmented family time.
How to use Continue's agent mode with Ollama for productive offline coding during your commute.
Build a semantic code history search using LanceDB and Continue's MCP integration to query your git history with natural language.
Breaking down what makes effective rules for AI coding agents, using Ruby TDD rules as an example.
Learn how to build fast, content-focused websites with Astro's unique approach to web development.
A conversation about the changes happening in DevRel.
This is a post to discover how to write a test for an expected error with Jest
Convincing your team to change up technologies can be challenging. Convincing my team to use GraphQL was not, mainly due to simplicity in getting a GraphQL prototype up and running using Apollo's Launchpad and our existing REST API.
Convincing your team to change up technologies can be challenging. Convincing my team to use GraphQL was not, mainly due to simplicity in getting a GraphQL prototype up and running using Apollo's Launchpad and our existing REST API.
podcast
Customize your bash with emojis
This is an example of the workflow at Netlify with react-storybook and how it cut down our development time between designer to developer.
I go through how to manage remote data-flow in a React Native application, using 5 of different techniques including starting with Promises and eventually getting to how I now got to GraphQL.
Queues are an abstract data structures when the collection of data are kept in order. Items can be enqueued to add to the queue and dequeued to be removed from the queue.
A Stack is an abstract data type that serves as a collection of elements with two principle operations, pop and push. Writing Stacks in Go is a great way to learn the language, which I will explain further in this post.
Go is one of the best languages to learn computer science. Thanks the compiler and built in fmt tool.
Injecting React into the asset pipeline is not so easy. Within the past 6 months I have discovered a great way to integrated React into my Rails app development workflow and take advantage of what everyone is excited about today.
Working with this can be confusing in React, especially when React has multiple ways to build components. Let's look the few ways to deal with this.
Compare Swift to React Native to find out whether or not it is worth the investment in learning iOS when you already know JavaScript
Learn how to implement SEO in your middleman site
DCI in Ruby
Deploying Ember with Divshot explained
How to get started with the rails-api gem
quick tips on learning Ember