AI/MLbeta

AI Code Review Assistant

A small prototype that turns a GitHub pull request into a contextual code review. I wanted to see how far you can push an LLM with a tight feedback loop, structured diffs, and a real linter running alongside. The result is a tool that catches the boring stuff (naming, dead branches, missing null checks) and surfaces the interesting stuff (design choices, hidden coupling) before a human even opens the PR. It is not a replacement for review, it is a sharper first pass.

OpenAI GPT-4 TypeScriptGitHub APINext.js

What this is

A lab, not a product.

Features

Learnings

Technologies

Capabilities

What it does

The features that actually got built and run in this prototype.

feature_01.ts

Diff-aware analysis that only reviews the lines that changed, with surrounding context windowed by AST

feature_02.ts

Security pattern detection that pairs with the API security playbook

feature_03.ts

Inline suggestions that map back to specific lines using GitHub PR review APIs

feature_04.ts

Style and best-practice nudges scoped to the project's own eslint config, not generic advice

feature_05.ts

Optional self-host mode so the model never sees code that should not leave your network

The stack

What it is built with

The libraries and runtimes I picked for this lab and why they earned their place.

OpenAI GPT-4 TypeScript

GitHub API

Next.js

What I learned

Learnings, in order of how much they surprised me

The things I would tell another engineer before they tried the same experiment.

Prompt engineering for code understanding is mostly context management, not clever prompts. See the shipping AI features playbook

Chunking strategies matter more than model size for large files. A 1k-line file split poorly gives worse reviews than a 200-line file with the right neighbours

Combining static analysis (the boring linter) with LLM judgement (the interesting questions) is much better than either on its own

Hallucinated bugs are real and expensive. The model needs a structured way to say 'I am not sure', otherwise reviewers learn to ignore it

Note: This is an experimental project in the beta stage. It is a learning exercise and technical exploration rather than a production-ready solution. Patterns and code may change.

AI/ML

Related labs

Other explorations in this area.

alpha

AI Form Generator

You describe a form in plain English, the tool generates a fully validated React component with proper field types, accessibility, and a Zod schema for validation. I wanted to test how reliable structured LLM output has become in 2025-2026 for a tightly scoped generation task. The answer is: very reliable, as long as you constrain the output schema aggressively and validate the model's response before rendering it.

experimental

Voice-First Interface

An experiment in voice-driven web UIs with real-time transcription and natural language commands. The trigger was watching how often I reach for keyboard shortcuts in tools I use daily, and wondering whether voice could be a faster path for some of those interactions. The answer is: sometimes, on the right device, in the right room. This prototype combines Whisper transcription with an intent classifier and a small command router. It is interesting, not a product.

experimental

AI Testing Agent

An AI agent that writes and runs end-to-end tests by reading your app like a user would. It opens pages with Playwright, identifies elements visually and structurally, drafts test cases, and runs them. The honest result: it is great for happy paths and visual regression, mediocre for complex flows, and useless for anything that requires domain knowledge it does not have. Worth it as a first sweep, not as your test suite.

Want me to build something like this for you?

If this kind of work fits your roadmap, I take on a small number of paid projects each quarter.

Start a project Just say hello

Real-time Collaborative Editor