Member of Technical Staff building production speech and language models for voice AI agents in financial services. Own core modeling, evals, and deployment on a small team with high autonomy and real revenue impact.
200k – 300k/yr
On-siteEntry levelML Engineering
About the role
What You’ll Do
Build and improve production speech models and systems with the ultimate goal of improving real-time conversational quality.
Post-train LLMs for voice agents used for consumer finance.
Develop evals, harnesses, and monitoring systems.
Work across models, data, infrastructure, and product to improve end-to-end agent quality.
Own major technical areas with significant autonomy, from problem definition to production deployment.
What We’re Looking For
Strong engineering fundamentals and ability to build reliable production systems.
Experience with speech modeling, or LLM post-training.
Familiarity with areas such as ASR, TTS, turn detection, speech enhancement, multilingual speech, LLM post-training, or model evaluation.
Ability to diagnose messy real-world model failures and turn them into practical improvements.
High ownership, strong execution speed, and comfort operating in ambiguous technical areas.
Prior experience in speech modeling and LLM post-training is ideal, but not required. We care more about slope, intensity, and technical ability than exact background. Exceptional engineers who are highly driven and excited to work on hard AI problems are encouraged to apply!
Why Join
Work on the full stack of production voice AI.
Own and lead core product areas on a small, high-caliber team.
Ship improvements that affect millions of live conversations.
Tackle real-world speech problems that are still unsolved outside of benchmarks.
Collaborate in person 4 days a week in our San Francisco office with a start time of 8:00 AM.
Benefits: medical, dental, and vision coverage, a generous 401(k), and catered lunches.
Skills
Speech ModelingLlm Post-TrainingAsrTtsSpeech EnhancementModel EvaluationProduction SystemsEvals And MonitoringPythonMachine Learning
Research Engineer building platforms to customize open-source LLMs via fine-tuning, RL, and evaluation. Focus on integrating post-training with inference engines (vLLM, SGLang, TensorRT-LLM), optimizing for RL workloads, and ensuring production reliability. Requires 2+ years ML production experience and strong Python/Go skills.
Builds evaluation pipelines, LLM judges, and observability tools to measure and improve AI assistant quality. Requires 2+ years software engineering with Go/Python, LLM eval experience, and analytical rigor for backend ML infrastructure.
200k – 300k/yr
Hybrid2+ YOEML Engineering
Founding Engineer
ProxSan Francisco, CA
Founding Engineer builds knowledge engines, multimodal agents, voice AI, and codegen systems for complex physical product support. Owns end-to-end customer deployments in fast-paced startup environment.
Build evaluation pipelines, LLM judges, and observability tools to measure and improve AI assistant quality. Requires 2+ years software engineering with strong Python/Go skills and LLM eval experience.
200k – 300k/yr
Hybrid2+ YOEML Engineering
Environment Engineers
AfterQuerySan Francisco, CA
Designs datasets, evaluation rubrics, and reward signals for RLHF/RLVR pipelines to expose model failure modes and improve frontier AI capabilities. Partners with AI labs; requires 1-4 YOE and passion for data-driven model behavior.