Published lab notes only.
KyaniteLabs publishes notes after there is a real build, lesson, product, or workflow to explain. The blog covers open-source AI tools, MCP systems, agentic media, developer learning, implementation notes, and the work behind the proof.
A routing policy for the actual GPT-6 lineup: Sol for discovery and coding, Luna for bounded high-volume processing, Astra when nothing else holds — each with a verification contract, and the lanes where a local 27B still beats all three.
The Equalizer Bench: a 3B that can't write ffmpeg, the same 3B shipping video edits, and the bug our own benchmark caughtWe benchmarked our own thesis: tiny model + deterministic guardrail layer vs raw capability. The curve is textbook, the trim trap is real, and the bench indicted our own product before anyone else could.
The One-Line Bug That Crashed Our Fast Lane: finding, fixing, and measuring a speculative-decoding crash on a $1,400 mini-PCSpec decoding crashed our fastest lane on day one. The bug was one missing line in our fork; the fix bought +17.7% and a surprise about draft quantization.
Two models, one $1,400 mini-PC: the paired numbers, failures includedA 35B reasoning model now runs shoulder to shoulder with our daily 27B on one $1,400 box, at the same time. Every number paired, same problems, same machine. The failures are in here too.
How I became a forward deployed engineer without a software engineer titleThe title is new; the work is old. Twelve years of enterprise deployments plus public, measured AI work. The honest path, artifacts included.
Evals are the FDE skill nobody lists: my 495-trial public benchmarkThe market says it cannot find people who can build AI evals. The skill is learnable and I published mine: 495 trials, certified floors, sabotage cell, open source.
Forward deployed vs solutions engineer vs implementation vs customer engineer: the title decoderFour titles, one job family, different coding bars. A decoder that reads any posting and tells you what you are actually signing up for.
What does a forward deployed engineer actually do? A demo on a $1,400 mini-PCThe straight answer, then the receipts: the whole FDE job run end to end on a $1,400 mini-PC, with public evals.
The Delegation Card: we asked a $1,400 mini-PC to take our jobsNot is-it-smart but can-you-hand-it-work-and-walk-away. 495 certified trials, then re-validated at deeper n after the product changed: 965 total, floors to 92.8%.
Qwen 3.8 27B on Strix Halo: the complete measured storyEvery dial measured, every number public: the frozen optimal config for a 27B on a $1,400 mini-PC.
Lab Notes: the measured-knees method for reasoning effortA methods note on reasoning-effort calibration. Thinking rescued 15/40 hards vs 4/40 off. On easy tasks it bought nothing. Publish the knee.
Lab Notes: 67% LiveCodeBench-30 on a $1,400 rig20/30 = 67% LiveCodeBench-30 on a $1,400 mini-PC. Wilson 95% CI 49-81%. Easy 10/10, medium 8/10, hard 2/10. n=30 public subset. Not the card.
Lab Notes: we reverted a llama.cpp regressionWe found a llama.cpp regression and reverted it. n=6 battery on a $1,400 rig: 0/6 slashes before, 5/6 after. Same Q4_K_XL. Not a quant story.
Lab Notes: 93% HumanEval on a $1,400 rig28/30 = 93% HumanEval on a $1,400 mini-PC. Qwen3.8-27B Q4_K_XL. Temp 0, thinking off. Failures: 50 and 145. Raw log linked.
Lab Notes: the model that can't forget but can't rememberThe number first: this model is 75% not a transformer. 48 of 64 layers keep a running state. On a $1,400 rig, load 198k once (1818s), then query in 9-27s.
Lab Notes: the basin was a bugWe published a basin. The serving binary was the hole. After the c7d8722 revert: 6/6 HIT at 198k. n=1 map of the fixed build.
Lab Notes: the KV verdictThe q8-to-q4 KV question from night 1 got its paired answer: same accuracy, zero tripwires, half the cache. Plus the label we had to correct in public when the server's counter beat our estimate.
Lab Notes: verdict nightA $1,400 mini-PC serving a 27B at 262k context asks one question: does capping thinking cost accuracy? The paired answer, the invalid first attempt it survived, and the doctrine that followed.
One mini-PC, one night, and the numbers that argued with themselves: tuning Qwen3.8-27B on Strix HaloA fully measured night-and-evening of tuning a 27B dense model on a Strix Halo mini-PC: the honest bands, the reversal, the crash doctrine, and the EC detective story.
Agents need verifiable tools, not better prompt theaterThe useful agent pattern is not a prettier prompt. It is a tool surface the agent can call, inspect, verify, and revise.
Repo history is a product signalA repo is not just storage. It is evidence of decisions, repairs, release behavior, naming drift, test gaps, and what the builder actually knows how to finish.
Implementation help is part of the product surfaceA useful open-source tool still needs a path from public repo to working environment. That path is product work, not an afterthought.
Why Kinocut mattersKinocut gives AI agents callable handles on timelines, effects, Hyperframes, and finished media at kinocut.dev.
Infinite monkeys, LLMs, and the room around the machineThe argument behind the video: output quality is not just probability. It is architecture, filters, and human taste.
What a working AI tool needs before people can use itA practical checklist for turning a working tool, workflow, or rough app into something other people can understand, install, and use.
MCP server implementation checklistThe checklist Kyanite uses to decide whether an MCP server is a toy, a usable tool, or something worth implementing.
Repo archaeology turns history into proofWhy commit history is one of the strongest proof sources for learning diagnostics, implementation help, and engineering trust.
AI discovery needs more than a sitemapWhat Kyanite adds so search engines and AI assistants can understand the tools, products, proof, and support path.
The repos are proof before the pitch.
Public repositories show what Kyanite builds, learns, breaks, fixes, and releases. The paid path helps people get those tools working in real environments.
