Case studies — maitre999

Approach: We built a Rust code-intelligence engine that indexes a repository into a symbol graph (redb) and a full-text search index (Tantivy), then serves navigation, impact analysis and security audits over a single MCP tool surface any agent can call.
Client: In-house

codenav: code intelligence instead of code search

Outcome

codenav now runs indexing daily across our own toolchain: reuse-before-write is a standing rule, refactors ship with a measured blast radius, and the same engine backs the code intelligence in our client work.

Approach: We measured the real constraint first: CPU, iGPU and NPU all share one 120 GB/s memory bus, so decode speed is a bandwidth budget, not a raw compute one. We pinned clocks, benchmarked model and quantization combinations against that budget, and wired a model router with memory guardrails so only one model ever holds the GPU at a time.
Client: In-house

The EVO-X1: one box, three compute engines, no cloud

delivered

Outcome

The box now hosts and builds itself: a coding agent, a knowledge base, and the render pipeline for this site, all running local, revertible, and reachable only over a private Tailscale mesh — no standing cloud dependency for the work that matters most.