Projects
Code to read, simulations to play, tools to try, a model that runs in your browser, and measured comparisons.
Open source
5 projects
Code you can read, run and check.
Open source
firstwatch
Your agent reads web pages, files and tool output, and anyone can hide instructions there. firstwatch scans for them before the agent acts, and makes it ask you first.
Open source
Agent skills
Skills you can drop into your coding agent: find Upwork jobs that fit you, turn an article into a video, free RAM, close a session cleanly. Install one, or all with one command.
Open source
doceval
Point it at your extractor and a labeled dataset. Get field-level accuracy, a failure taxonomy and per-document cost, without writing eval infrastructure.
Open source
mcplytics
You put an MCP server online and AI apps call it. Which tools do they use, which calls fail, and how fast is it? One wrapper answers that, with no backend.
Open source
tally-aiagent
Tally invents any master name it doesn't recognise, then reports success. This refuses names the books don't hold, and verifies every entry after posting.
Simulations
2 simulations
Attack an AI agent or defend it, then see what it did and why.
Simulation
The Butler
An AI butler tidies an inbox with real tool calls. Attack it with prompt injection, defend it with rules, locks and approvals, then read the score and the trace.
Simulation
Plug
An AI agent does one job through MCP servers. Turn a server bad, then connect only what the job needs and lock the rest. See the JSON-RPC and the trace.
Live tools
3 tools
Try them in your browser.
Live tool
Play with Jev
Games and toys built on Jev, an AI that answers in probabilities, not words. Race it to find Pip, ask the Magic Jev Ball, or let waif name a feeling.
Live tool
affterms
Affiliate terms for 471 dev and SaaS tools, each figure sourced and dated. A calibrated model says which pages changed; a person reads them before any figure moves.
Live tool
India Facts
Ask a question of fact about India. An answer with no opinions, and every sentence links to the record it came from.
Models
1 model
Small models I trained, each with a playground that runs in your browser.
Model
tinytowns
A very small AI model that makes up town names. Type some letters and see its guesses for the next letter. It does not know which places are real.
Benchmarks
3 benchmarks
Measured comparisons from real builds: accuracy, speed and cost.
Benchmark · 14 Oct 2026
Small model vs LLM
One small job, five kinds of tool: a 4 MB model, open language models, Claude and a plain list. What each one is good at, and what it costs.
Benchmark · 12 Feb 2025
GPT-4o vs. Veryfi Extraction Benchmark
GPT-4o against Veryfi on irregular financial scans, across several document types — where each one wins, and what each extraction costs.
Benchmark · 31 May 2026
Self-Hosted Whisper Benchmark
A self-hosted meeting-transcription pipeline, plus a benchmark of Whisper model sizes on a real 32-minute recording to find the accuracy/speed sweet spot.