Source Job

$250,000–$400,000/yr
North America

  • Build reinforcement learning environments for complex clinical reasoning tasks.
  • Design and calibrate reward functions, grading rubrics, and LLM judges.
  • Collaborate with researchers and clinicians to validate model outputs and improve environments.

Python PyTorch Transformers Reinforcement Learning

20 jobs similar to Reinforcement Learning Engineer

Jobs ranked by similarity.

$250,000–$400,000/yr
Canada United States

  • Build reinforcement learning environments for multi-step clinical reasoning, including tool use and success criteria.
  • Design and calibrate reward functions, grading rubrics, and language model judges to prevent reward hacking.
  • Own the full loop from task design through evaluation and delivery to model-training teams.

The company builds decision-support tools for clinical care, using reinforcement learning to enhance clinical reasoning. It is a healthtech team that collaborates with clinical experts to improve model outputs and develop high-quality clinical AI systems.

US

  • Build, train, and evaluate machine learning models, including LLM fine-tuning with supervised and reinforcement learning.
  • Design RL environments, reward functions, and evaluation frameworks; diagnose training issues and document results.
  • Collaborate with domain experts to structure knowledge and present findings clearly.

$250,000–$400,000/yr
North America

  • Build and improve pipelines that parse clinical documents into validated, executable modules.
  • Run experiments across open and closed models, fine-tuning, and constrained decoding methods.
  • Work with clinicians to evaluate outputs and turn model failures into system improvements.

They build a compiler that converts clinical standards into verified, traceable decision logic for patient care. This healthtech team combines frontier model research with production engineering to improve clinical decision support.

$250,000–$400,000/yr
North America

  • Build production pipelines that parse clinical standards into executable modules using document and layout models.
  • Run structured experiments comparing open and closed models, fine-tuning, and constrained decoding for guideline logic.
  • Create deterministic validators and benchmarks to surface ambiguity, contradictions, and missing evidence in clinical outputs.

They build a production compiler that turns clinical standards into verified, traceable decision artifacts for healthcare. The team is research-focused and collaborative, with clinicians and technical leadership working closely; company size is not disclosed.

$250,000–$400,000/yr
Canada US

  • Build pipeline components that parse clinical documents and transform content into executable modules.
  • Design and refine a target syntax for guideline logic, and evaluate open and closed models with benchmarks and adversarial cases.
  • Develop deterministic validators using static analysis or formal methods to surface ambiguity and contradictions.

$129,750–$144,000/yr
Canada 6w PTO

  • Own the AI that assesses patients for US mental healthcare, from conversation to treatment recommendation.
  • Build evaluation harnesses and simulated patient conversations to test every configuration before deployment.
  • Raise the engineering bar through architecture, code review, and mentorship while ensuring regulatory compliance.

Limbic develops AI mental health chatbots for patient assessment and therapy, with clinical validation and FDA/NHS adoption. The company has supported over 820,000 patients and emphasizes safety, science, and an inclusive culture.

Global

  • Conduct research on reinforcement learning algorithms for multimodal models, including diffusion-based approaches and unified frameworks.
  • Design and build scalable RL infrastructure and reward modeling to support distributed training and improve stability.
  • Develop benchmarking protocols, curate multimodal datasets, and advance next-generation RL paradigms for real-world applications.

Tether is a fintech company pioneering digital finance solutions, including the stablecoin USDT, and developing energy, data, and education products. The company is a fast, lean global team of top talent working remotely across the world, with a culture focused on innovation and transparency.

Global

  • Own applied RL programs from ambiguous customer problem to deployed improvement loop.
  • Understand workflows, build environments and evaluations, run post-training experiments, and deploy results.
  • Work directly with customer researchers, engineers, and leaders, accountable for research result and practical impact.

Prime Intellect builds the open superintelligence stack, providing a full-stack platform for frontier-scale post-training of AI models. The company has raised $150M from top investors and is seeking builders who want to work at the intersection of frontier research, real infrastructure, and go-to-market.

$150,000–$300,000/yr
Global

  • Build post-training and evaluation systems for multimodal agents operating browsers and desktops.
  • Turn real workflows into environments, rewards, and experiments that drive reliable agent improvement.
  • Work directly with customers and research partners while maintaining a substantive research agenda.

Prime Intellect is building the open superintelligence stack, providing infrastructure for post-training frontier AI models. The company has raised $150M in funding and fosters a culture of ownership and collaboration for ambitious AI teams.

$50–$50/hr
US Canada

  • Contribute to post-training and evaluation infrastructure for a novel multimodal AI system.
  • Fine-tune models using supervised and reinforcement learning approaches (e.g., SFT, RLAIF, RLVR).
  • Define custom metrics and develop tools to assess and mitigate AI-related risks in generative AI.

Sony AI is Sony's research organization using AI to unleash human creativity, working with gaming, entertainment, and music divisions. The Ethical Creative and Protective AI Lab operates as a globally distributed team across the U.S. and Canada, with some 900 million Sony devices in use worldwide.

$130,000–$200,000/yr
US

  • Develop AI methods and algorithms for reliable agents.
  • Design simulations and evaluations for agent behavior.
  • Partner with product and engineering to ship research.

The company builds an AI agent platform focused on making agents reliable in production. It is a research-driven environment with a collaborative, startup-oriented culture.

$73,150–$114,950/yr
Europe

  • Own ML projects end-to-end, from exploration to production deployment, iterating based on real user feedback.
  • Build agentic LLM systems with multi-step workflows, tool use, retrieval, and orchestration for reliable clinical use.
  • Treat evaluation as core engineering work, building eval sets, LLM-as-judge pipelines, and regression tests to catch quality drops.

Sword is building AI to revolutionize healthcare, focusing on clinical reasoning and real-time treatment. With over 700,000 members and 1,000+ enterprise clients, they are backed by $500M+ from top investors like Khosla Ventures and General Catalyst, defining a new standard in AI-native care.

$250,000–$400,000/yr
North America

  • Build and improve pipelines for parsing clinical documents and transforming guideline content into executable modules.
  • Evaluate open and closed models, fine-tuning approaches, and constrained decoding strategies for clinical fidelity.
  • Develop deterministic validators and benchmarks to assess correctness and expose ambiguities in clinical logic.

They build a production compiler that turns clinical standards into structured, executable decision logic. The team is research-focused and collaborative, working across engineering and clinical disciplines.

$150,000–$300,000/yr
Global

  • Build and post-train agents for scientific research, including literature search, code execution, and domain tool use.
  • Develop RL and multi-agent methods for proposing, testing, and critiquing hypotheses, with strong baselines and evaluation.
  • Design verifiers, reward signals, and reproducible experimental loops for scientific workflows, ensuring evidence-based improvements.

Prime Intellect builds the open superintelligence stack, providing infrastructure for frontier AI labs to train, evaluate, and deploy models at scale. The company has raised $150M from top investors and operators, and is building a culture focused on frontier research, real infrastructure, and go-to-market for a new category.

Global

  • Maintain and improve the platforms by fixing bugs, paying down technical debt, and improving test coverage and documentation.
  • Build new features across the stack, from database and backend to APIs and internal tools.
  • Build RL environments and the infrastructure to run them at scale, including sandboxes, environment images, rollouts, and reward computation.

The Human Data Division builds training and evaluation data for frontier AI labs. We are a small team that moves fast and holds a high quality bar.

$250,000–$400,000/yr
North America

  • Build production pipelines that parse clinical standards into validated, executable modules.
  • Design and refine representations for guideline logic, including exclusions, precedence, and recommendations.
  • Evaluate models, develop benchmarks and adversarial cases, and ship successful improvements into production.

This healthtech company builds a production compiler that turns clinical standards into executable, traceable decision logic. The team is research-focused and collaborative, working across disciplines to improve the reliability of clinical decision systems.

Global

  • Build quantitative-research environments, agents, and post-training methods for systematic decision-making.
  • Turn workflows into tasks, run experiments against real baselines, and deploy what works.
  • Work directly with customers and quantitative funds while maintaining a substantive research agenda.

Prime Intellect builds the open superintelligence stack for post-training AI, unifying compute, environments, evaluations, and deployment. They have raised $150M from top investors and are building a culture of ambitious, externally-facing applied research.

$130,000–$200,000/yr
US

  • Develop and improve methods, algorithms, and systems across the lifecycle of reliable AI agents.
  • Create new approaches for simulating, evaluating, and optimizing agent behavior in real-world settings.
  • Partner with product and engineering teams to bring AI research into practical applications.

The company builds infrastructure for reliable AI agents in production, targeting enterprise workflows. It is a focused team of researchers and engineers working collaboratively on advancing agentic AI.

Global

  • Design and build AI conversations and agents for advisory experiences, practice simulations, and diagnostic coaching.
  • Turn scenario specs, decision models, and assessment rubrics into prompts, application logic, and testable behavior.
  • Implement grounding and retrieval from approved knowledge sources, with source references where required.

DistantJob is at the forefront of learning technologies, providing e-learning programs for professionals in regulated industries. They run efficient Agile-first teams, have a broadly distributed staff, and value good-natured, inquisitive, and methodical IT professionals who collaborate and share knowledge.

$100,000–$225,000/yr
Global

  • Work with product, design, engineering, research, and go-to-market partners to identify user friction and prioritize improvements.
  • Build frontend and backend features for onboarding, environment submission, review, and discovery workflows.
  • Design interfaces that help users understand agent runs, evaluation results, reward signals, and quality feedback.

They are building a platform and marketplace for reinforcement learning environments, serving AI research teams. The company is remote-friendly with offices in San Francisco and Singapore and emphasizes user-centered product development.