24 oct
|
Kimchi
|
Valencia
Kimchi is seeking an applied ML Platform Engineer to own the routing model end to end: features, objective, training, calibration, and release cadence. You’ll treat it as a product, not a research artifact, shipping improvements with measurable impact.You’ll build production-parity evaluations, convert fleet data into training signals, and optimize the objective as task success × token cost × wall-clock speed under real concurrency. Remote-first & integral team.#J-18808-Ljbffr
📌 Remote Senior Applied Ml Engineer - Real-Time Llm Routing (Valencia)
🏢 Kimchi
📍 Valencia