Someone Hid a Creature Generator Inside Claude Code
We dug into the Claude Code binary and found a fully built deterministic companion system (species, rarity tiers, hats, stats) just waiting behind a feature flag. Happy April 1st.
From the workshop
Notes from the bench.
What we learn while building: AI agents, developer tools, and the occasional hard-won gotcha.
-
-
Why One Search Algorithm Is Never Enough: A Taxonomy of AI Memory Retrieval
BM25, HNSW, GraphRAG, temporal decay - a practical breakdown of the search paradigms powering AI memory systems, organized by the failure mode each one solves.
-
Prompt Caching: The 30% You're Leaving on the Table
Most developers optimize model choice, context size, and output quality, but never notice the 20-35% of their bill that disappears with one config flag. Here's what prompt caching actually is, how five major providers implement it differently, and what happens when your tooling silently breaks it.
-
I Benchmarked Every Embedding Model Worth Running. Here's What I'd Actually Deploy.
We went from nomic-embed-text to OpenAI's text-embedding-3-small and thought we'd upgraded. Turns out we'd moved from bad to mediocre. Here's the full landscape of self-hosted embedding models in 2026, organized by what you can actually run on your hardware.
-
Bigger Context Windows Don't Make Better AI
The arms race to 1M context windows assumes more context is better. Research from Chroma proved it makes things worse - and labs have been expanding context windows ever since.
-
Why We Built Chronicle: Code Search Designed for AI Agents
IDE search was built for humans. Grep was built for terminals. Neither was built for AI agents burning through context windows. So we built Chronicle - a persistent code index that gives agents structural understanding in ~50 tokens instead of 2,000.
-
Welcome to Tensaku Labs
Why we started this blog, what we build, and the iterative method behind everything we do.