Engineering the AI stack,
in the open.
Cuecoder is an AI engineering studio building hardcore developer tools and infrastructure for the AI era. What ships gets written about. What works gets open-sourced. The rest becomes product — all under one roof.
One CLI. From scaffold to ship.
Every Cuecoder product runs on the same opinionated toolchain — open-sourced as
cue
so any team can use it. Below: a real production deploy.
3
Products in production
$24.8k
MRR · Axon
12.4k
GitHub stars
4,218
Newsletter readers
Shipped, billed, and on-call for.
Axon
Eval & observability layer for AI agents in production. Trace every step, score every output, ship with confidence.
cue
Opinionated CLI for shipping AI features. Scaffolds, evals, deploys, observability — one binary, zero config.
Groundwire
Retrieval that actually grounds. Hybrid search, citation-first responses, and an eval harness that catches hallucination at PR time.
Field notes from shipping AI in production.
Evals are the product. Everything else is a side effect.
After two years shipping LLM features in production, the only artifact worth trusting is a well-designed eval suite. Here is how Cuecoder structures them.
Stop tuning prompts. Start tuning context.
Prompt engineering plateaus fast. The real lever is what you put in front of the model — retrieval, structured tools, and dynamic memory.
Why the gateway got rewritten in Rust
Two years of Python latency tax, a runaway autoscaler, and a 3am incident later — the case for moving the hot path off the GIL.