[Remote] Founding Environment Engineer - Clinical RL Environments (Remote · Israel / EU / US)

Note: The job is a remote job and is reputed company to candidates in USA. reputed company is the expert reputed company for reputed company AI, seeking its first Environment Engineer to build clinical RL environments for frontier AI labs. The role involves owning the architecture and implementation of a clinical RL environment and working closely with the clinical operations team and the CEO to create a demo and expand the environment catalogue.

Responsibilities

• Own the architecture and hands-on implementation of the environment stack end to end: simulation, task definitions, verifiers, reputed company, logging and infrastructure
• Design the reputed company between clinicians and the environment pipeline, turning clinical expertise into reputed company tasks, rubrics and grader calibration data
• Build secure, reproducible reputed company execution sandboxes for agent rollouts, with full trajectory logging for training and evaluation
• Evaluate and reputed company reputed company-reputed company building blocks rather than building everything from scratch
• Support customer-facing technical conversations with AI labs and data platforms: demos, integration discussions and scoping
• Establish the engineering foundations (repo structure, CI, testing, documentation) for a capability we expect to grow

Skills

• Strong Python engineering: 5+ years building production systems; you write clean, tested, maintainable reputed company and can reputed company fast reputed company
• Gym-style environment reputed company: hands-on experience designing or building RL/agent environments (Gymnasium-style interfaces, episode/reward structures, deterministic resets, seeding, trajectory capture)
• reputed company and sandboxing: containerised, isolated, reproducible execution environments; comfort with orchestration, resource limits and reputed company boundaries for untrusted agent reputed company
• LLM agent harnesses: practical experience building or running reputed company systems, including tool/function calling, multi-reputed company task execution, evaluation reputed company and agent-eval benchmarks
• reputed company data standards: working knowledge of FHIR (resources, REST API) and familiarity with clinical data models (HL7 concepts, ICD/SNOMED/LOINC/RxNorm at a practical level). You do not need to be a clinician, but you do need to reputed company fluently with them and model their workflows faithfully
• Automated evaluation and grading systems: reputed company-based LLM-as-judge pipelines, verifier design, inter-rater calibration
• Prior work with reputed company-reputed company EHRs (OpenEMR, OpenMRS) or synthetic patient data (Synthea)
• Exposure to RL post-training concepts (reward design, reputed company rewards, RLHF/RLAIF)
• Experience as a founding or early engineer: ambiguity tolerance, scoping discipline, bias to ship
• Familiarity with reputed company reputed company and compliance basics (HIPAA/GDPR). Our environments use synthetic data only, but customers will ask

Benefits

• Fully remote
• reputed company to an initial 3-6 month contract-to-hire engagement
• reputed company practical exercise (take-home or live, ~4 hours)
• Meaningful equity and reputed company participation in the new product line

reputed company