Senior Applied ML Engineer | Kimchi (Model Router)

The role

You will own the routing model end to end, including features, objective, training, calibration, and release cadence. You will build evaluations to gate changes and turn fleet data into training signals to improve the router.

What you will do

• Own the routing model end to end including features, objective, training, calibration, and release cadence.
• Build the evaluation that gates every change using production-parity benchmarks and router-vs-router comparisons.
• Turn fleet data into training signal from production outcomes and benchmark rollouts.
• Model the objective including task success, token cost, and wall-clock speed.
• Run the online loop including shadow, A/B, full rollout, and drift monitoring.
• Push the frontier with calibrated decisions, abstention under uncertainty, and speed-aware routing.
• Partner with platform engineers on instrumentation.

Required qualifications

• 5+ years in applied ML or ML research.
• Experience training, evaluating, and shipping models that make real-time decisions.
• Strong Python and SQL at scale.
• Knowledge of the difference between a model that got better and an eval that got easier.
• Opinions about what LLM benchmarks measure.
• Knowledge of serving reality including TTFT, decode speed, batching, prefix caching, and concurrency.
• Ability to instrument before optimizing with error bars.

Nice to have

• Worked on LLM routing specifically: RouterBench-class evals, cost/quality/latency Pareto fronts.
• Calibration, abstention, or conformal methods in production decision systems.
• Run online A/B loops with drift monitoring on a model that touches real money.

Benefits

• Competitive salary (depending on the level of experience).
• Flexible, remote-first global environment.
• Equity options.
• Quick feedback with a fast-paced workflow.
• 10% of work time on personal projects or self-improvement.
• Learning budget for professional and personal development.
• Annual hackathon.
• Team-building budget and company events.
• Equipment budget.
• Extra days off.

Before you apply

• Cast AI does not provide any form of visa sponsorship/work permit.
• A background check may be conducted at the final stage of recruitment through Checkr.

About the company

Cast AI is a Miami-headquartered, globally distributed team across 34 countries. It hit unicorn status in January 2026 with a $50B+ backed investment.