RL Research Engineer

Job Title: RL Research Engineer (Planning & Control)

Location: Vapi, Gujarat

Employment Type: Full-Time

Overview

We are seeking a highly skilled Reinforcement Learning (RL) Research Engineer specializing in planning and control. The role focuses on designing learning-based planners and policies (RL, imitation learning, model-based) and integrating them with classical control approaches to enable safe, efficient, and robust autonomous operation across multiple domains including humanoids, AGVs, cars, and drones.

Key Responsibilities

• Develop and train policies from human demonstrations and teleoperation data.
• Implement safe reinforcement learning approaches with constraints.
• Design long-horizon planners using world models and uncertainty-aware control.
• Implement safety shields, fallback controllers, and verify-before-deploy pipelines.
• Collaborate with cross-functional teams to integrate RL policies with control systems.
• Conduct sim-to-real transfer and ensure policies generalize in real-world settings.
• Design reward functions and implement offline RL and behavioral cloning strategies.

Must-Haves

• 4–8+ years of experience in RL and control systems.
• Strong expertise in Model Predictive Control (MPC), Control Barrier Functions (CBFs), reachability analysis, or similar methods.
• Master’s or PhD in Robotics, Control, AI, or a related field.
• Experience with sim-to-real transfer, reward design, offline RL, and behavioral cloning.

Nice-to-Haves

• Experience with multi-agent reinforcement learning.
• Knowledge of hierarchical options and diffusion policies.
• Familiarity with long-horizon task planning in complex environments.

Success Metrics

• Task success rate in target domains.
• Rate of human or system interventions during execution.
• Compliance with energy, jerk, and other control limits.
• Minimization of constraint violations in real-world deployment.

Domain Notes

Humanoids:

- Stable locomotion and bimanual task RL.

AGVs (Autonomous Ground Vehicles):

- Navigation in mixed human zones, traffic rule compliance, and aisle etiquette.

Cars:

- Interactive merges, handling unprotected turns, and safe navigation in dynamic traffic.

Drones:

- Wind-robust flight, safe landing and perching maneuvers.

Application Instructions

Interested candidates may apply by sending their resume and cover letter to [email protected] with the subject line: “RL Research Engineer (Planning & Control) Application”.