Job Description

Focus: Reinforcement Learning, AI Agents & Evaluation DeepRec.ai is partnering with a fast-growing AI company building technology at the intersection of reinforcement learning, AI agents and model evaluation . We’re looking for a Reinforcement Learning Engineer who enjoys building the systems and tooling that turn cutting-edge AI research into scalable, usable infrastructure. This is a highly hands-on role working closely with research and engineering teams. You’ll help build the platforms used to generate and evaluate agent behaviour, understand model performance, improve training data quality and accelerate experimentation. What you'll work on Build infrastructure and tooling supporting reinforcement learning and agent evaluation. Develop systems for collecting, inspecting and improving training and evaluation data. Create tools for analysing trajectories, rewards, model behaviour and failure modes. Build evaluation pipelines, quality-control workflows and experimentation tooling. Develop APIs and backend services connecting data generation, evaluation and training workflows. Create dashboards and observability systems for understanding experiments and model performance. Work closely with AI researchers to turn rapidly evolving research requirements into reliable engineering systems. Take ownership from initial idea through implementation, deployment and iteration. What we’re looking for You’ll likely have strong software engineering fundamentals alongside experience working with modern AI/ML systems. Core experience: Experience building production‑grade backend systems, platforms or developer tooling. Understanding of reinforcement learning, LLMs, AI agents or model evaluation. Experience with data pipelines, experimentation or ML infrastructure. Comfortable with Docker, cloud infrastructure, CI/CD and production observability. Ability to work effectively where requirements are evolving and problems aren’t fully defined. Experience with areas such as RL environments, reward systems, agent trajectories, evaluation frameworks, synthetic/training data, annotation platforms or AI research tooling would be particularly relevant. The environment You’ll join a small, highly technical team tackling difficult problems around the training and evaluation of increasingly capable AI systems. The environment is fast-moving and engineering-led. You’ll have significant ownership, work closely with researchers and other technical stakeholders, and be expected to build rather than simply advise. We’re particularly interested in engineers who enjoy open-ended technical problems, rapid experimentation and shipping systems that researchers actually use. #J-18808-Ljbffr

Job Tags

Similar Jobs