Reinforcement Learning Intern

About the company

Technoculture Research is developing micro-scale electrochemical laboratory systems designed to bring lab-quality diagnostics closer to everyday care. Its platform combines microfabricated electrodes, specialized surface chemistry, and microfluidics to run protein, nucleic-acid, and metabolite tests in minutes. By shifting from optical readouts to electron-based sensing, the company lowers both device and test costs substantially, helping make precision diagnostics more affordable and scalable across different care settings.

Role overview

The company is hiring a Reinforcement Learning Intern for its robotics team at SentientX (Sentient Industries Private Limited). This role focuses on RL-driven control policies for legged robots, including quadruped and humanoid platforms, and involves research as well as implementation work that connects simulated training with real-world robot behavior.

What you will work on

• Build, train, and assess reinforcement learning policies for robotic locomotion in simulation.
• Use physics-based simulators to design, test, and refine robot control strategies.
• Support research and implementation work around sim-to-real transfer.
• Develop and compare RL methods such as PPO and SAC for locomotion use cases.
• Record experimental setups, outcomes, and research insights in a clear and structured way.
• Coordinate with the robotics engineering team on research and deployment tasks.

Requirements

Candidates should have a strong grasp of reinforcement learning methods such as PPO, SAC, or similar approaches, along with solid Python programming skills and hands-on use of RL libraries such as Stable-Baselines3, RLlib, or comparable tools. Familiarity with physics-based simulation platforms like MuJoCo, Isaac Gym, PyBullet, or related environments is important, as is comfort with Git and collaborative development workflows. A background in robotics, control systems, or mechanical engineering is preferred. Experience with locomotion training, quadruped or humanoid robots, URDF, or ROS is an added advantage. Prior research or internship exposure in robotics or machine learning labs is also valued. Applicants should be highly self-driven, curious, and able to work effectively in a fast-moving research setting. A GitHub repository, project, or assignment demonstrating practical RL implementation is required. Strong understanding of sim-to-real transfer, domain randomization, system identification, and robust policy training is expected.

Eligibility

This opportunity is open to candidates who can commit to a full-time, in-office internship, can begin between 29 June 2026 and 3 August 2026, can stay for 6 months, and have relevant skills and interest. It is specifically suited to students currently pursuing Computer Science Engineering.

Perks and benefits

• Certificate on completion
• Informal dress code
• Free snacks and beverages
• Hands-on exposure to advanced robotics and Physical AI research

Additional details

The internship stipend is INR 18,000 to 22,000 per month. The application deadline is 29 July 2026 at 23:59:59. The role is based in Bangalore, Karnataka, India, with one opening available.