[Remote] Reinforcement Learning Engineer

Note: The job is a remote job and is open to candidates in USA. Bright Vision Technologies is a forward-thinking software development company dedicated to building innovative solutions that help businesses automate and optimize their operations. They are seeking a skilled Reinforcement Learning Engineer to design, train, and deploy RL-based systems for high-impact decision-making problems where supervised learning alone is insufficient.ResponsibilitiesDesign and implement reinforcement learning solutions for sequential decision-making problems in real and simulated environmentsDevelop, calibrate, and maintain simulation environments suitable for large-scale agent trainingImplement and evaluate modern RL algorithms including policy gradient, actor-critic, off-policy, and offline RL methodsEngineer reward functions and shaping strategies that align agent behavior with desired outcomes and safety constraintsApply offline RL and imitation learning techniques where exploration is costly or unsafeUse RLHF, DPO, and related techniques for fine-tuning large language models when relevantBuild scalable training infrastructure for distributed RL, including efficient experience collection and replay systemsOptimize training stability and sample efficiency through algorithmic and engineering improvementsDesign rigorous evaluation protocols, including out-of-distribution and adversarial test casesImplement safety mechanisms such as constraint enforcement, conservative policies, and human-in-the-loop oversightCollaborate with applied scientists and product teams to identify high-value RL use casesMonitor deployed policies and models in production for drift, regression, and unintended behaviors, building the alerting and dashboards that surface issues before they meaningfully affect usersDocument methodology, design decisions, and operational characteristics for internal stakeholdersStay current with RL research and translate promising techniques into production-ready solutionsSkillsMaster's or PhD in Computer Science, Machine Learning, or a related field; or equivalent applied experienceSix or more years of combined RL research and engineering experienceStrong proficiency in Python and modern deep learning frameworksHands-on experience with at least one major RL library or in-house RL stackSolid understanding of probability, optimization, and the theoretical foundations of RLExperience designing and tuning reward functions in non-trivial environmentsFamiliarity with simulation environments and large-scale experience collectionExperience training neural network policies on GPU clustersStrong written and verbal communication skillsTrack record of shipping or publishing impactful RL workExperience with RLHF for large language modelsFamiliarity with multi-agent RL or hierarchical RLExposure to robotics, control systems, or autonomous drivingPublications in RL or related research venuesOpen-source contributions to RL libraries or environmentsBenefitsCompetitive base salary commensurate with experience, plus benefits.Full-time, direct W2 with Bright Vision Technologies (no C2C, no 1099, no third-party)100% remoteLong-term, multi-year, aligned to the Bright Vision SOW delivery roadmapWe will support H1B transfers for qualified candidates.Company OverviewBright Vision Technologies is an information technology company that offers software development, AI, and cybersecurity services. It was founded in 2020, and is headquartered in Bridgewater, New Jersey, USA, with a workforce of 51-200 employees. Its website is https://bvteck.com.