May 23, 2026 Trying DuckDB With an AI Query Chat Evidence.dev runs on DuckDB-WASM in the browser. I wanted to try DuckDB directly, so I built a chat that asks Claude for SQL and runs it locally.
duckdbwasmsqlllm
Read article
ilo: A Programming Language for AI Agents, Not HumansMay 22, 2026 Three Bets on Long-Context Attention Gemma 4, Qwen 3.6, and DeepSeek V3 each take a different path to long context. A reverse-engineering test shows where the trade hurts.
llmaiattentionmodel-comparison
Read article
LLM vs JEPAMay 16, 2026 Part 2: Why Pixel Prediction Goes Blurry Yann LeCun says generative models fail on video because the future is ambiguous. I tested the claim with a 700K-parameter PyTorch model on a synthetic bouncing ball.
llmmachine-learningpythonpytorchcomputer-visionjepaself-supervised-learning
Read article
LLM vs JEPAMay 15, 2026 Part 1: Yann LeCun's Bet Against LLMs The LeCun argument in plain terms, with the papers behind it. The map for a five-part series testing his recommendations on a MacBook.
llmmachine-learningaijepaworld-modelsself-supervised-learning
Read article
Token Efficiency: Where the Tokens GoMay 14, 2026 Decode throughput is bandwidth divided by model size Single-token decode reads every weight once, so the spec sheet predicts tokens per second before you download anything. Measured against six Qwen models on an M3 Max.
llmai
Read article