I'm a software engineer - well I prefer the term writer.
I have an eclectic background working in everything from self-driving cars to SaaS software. Backend, frontend, machine learning, deep learning, database munging, etc.
I've also done a fair bit of open source, where I most notably created the formatter for the Julia programming language. Most of what I work on in my spare time is some combination of Julia and tinkering with LLMs.
This site consists of technical writings, typically pertaining to a coding project. My substack consists whatever else I feel like being judged for.
An GPT powered text to code generator and execution environment which allows you to transform, query, and visualize data in a chat-like interface. Try it out!
A width-aware formatter for Julia code. Quite possibly the most customizable formatter, for any language out there.
Generates Julia types and functions from a GraphQL schema.
A research assistant for Anthropic AI technical writings. I made this for myself while I was studying the transformers circuits series. Frontend is made with SvelteKit and TailwindCSS, hosted on Vercel. Backend is hosted on Railway and uses embeddings to extract relevant information from posts which the LLM uses to answer questions.
A experiment in editing tables (dataframes) data in the browser. Built with SvelteKit and TailwindCSS. Using Pyodide (WASM) to run Python in the browser. The user describes edits they want to make to the data and the LLM generates Python code, which is then executed in the browser and data is updated in realtime. Additionally, the AI will explain what it did.