Founding RL Engineer - Healthcare AI & Clinical Reasoning

BioStack Platforms, Inc.; San Francisco (CA); Sur place; USD 200 000 - 350 000; Plein temps

Avantages offerts par ce poste

Equity 0.1%-1.0%

401(k) match

Health benefits

Meals

Unlimited AI tools budget

Relocation & joining bonus

Visa sponsorship available

Résumé du poste

BioStack Platforms, Inc. is seeking a Founding Research Engineer, RL/Reasoning, to build reinforcement learning environments and post-training systems for healthcare AI.

The role focuses on turning real-world clinical data into model-ready tasks, rewards, verifiers, and agent environments to enable learning from meaningful outcomes. The candidate will work across the full RL loop—from environment design to training, evaluation, and iteration—addressing clinical reasoning, longitudinal patient

Qualifications

• Experience with reinforcement learning, post-training, agents or related ML systems.
• Ability to design environments and rewards for RL tasks.
• Strong judgment around data quality, coverage, and clinical relevance for training and evaluation.

Responsabilités

• Build healthcare-specific RL environments, including tasks, action spaces, reward functions, verifiers, and evaluation harnesses.
• Run post-training experiments on language models and agents (SFT, RLVR, RLHF/RLAIF, and reward modeling).
• Turn clinical/biomedical datasets into training environments with measurable outcomes.
• Design rewards and verifiers that reflect clinical reasoning and longitudinal decision-making tasks.
• Train and evaluate multi-step agents across patient histories and structured medical data.
• Build scalable pipelines for rollouts, training, evaluation, and dataset iteration.
• Analyze model failures to improve environments, rewards, datasets, and training runs.

Connaissances

Reinforcement learning

RL post-training

Agent environments

ML systems

Description du poste

BioStack Platforms, Inc. is seeking a Founding Research Engineer, RL/Reasoning, to build reinforcement learning environments and post-training systems for healthcare AI.

The role focuses on turning real-world clinical data into model-ready tasks, rewards, verifiers, and agent environments to enable learning from meaningful outcomes. The candidate will work across the full RL loop—from environment design to training, evaluation, and iteration—addressing clinical reasoning, longitudinal patient