Manipulation Capabilities Engineer (Robotics)

TL;DR

Manipulation Capabilities Engineer (Robotics): Teaching hirify.global robots to manipulate objects and perform real-world tasks with an accent on deep learning, reinforcement learning, teleoperation, and sim-to-real transfer. Focus on building the full data-to-deployment loop, improving manipulation policies, expanding robot capabilities, and solving control and hardware integration challenges.

Location: On-site in London, UK

Company

Builds commercially scalable and safe hirify.global robots for industrial applications, including the HMND-01 Alpha platform.

What you will do

• Post-train manipulation policies using behaviour cloning and reinforcement learning, owning the process from data collection through deployment.
• Develop data preprocessing strategies and improve the quality, diversity, and coverage of collected data.
• Set up reinforcement learning in digital-twin simulations and optimize rewards and simulation quality for reliable transfer to real robots.
• Define data requirements for specific capabilities and collaborate with data collection and teleoperations teams.
• Expand robot observation and action spaces, expose new components to operators, and improve motion smoothness and teleoperation workflows.
• Work with controls and hardware design teams to incorporate manipulation findings into future robot generations.

Requirements

• At least 3 years of experience working with robots in industry or research, with shipped artifacts to demonstrate impact.
• Strong understanding of modern teleoperation and low-level control stacks, including experience diagnosing real robot hardware issues.
• Experience with neural network post-training and applied deep learning, including data curation and policy fine-tuning.
• Familiarity with streaming datasets, checkpointing, state management, distributed training, and PyTorch or JAX.
• Ability to profile and debug numerical issues, write maintainable research code, document experiments, and communicate trade-offs clearly.
• Good understanding of modern software engineering practices.

Nice to have

• Experience training vision-language-action models for manipulation, including autoregressive, diffusion, or flow-matching approaches.
• Familiarity with OpenVLA, Physical Intelligence models, or similar open VLA frameworks.
• Experience applying reinforcement learning to robotics problems.
• Top-tier robotics or deep learning publications, or equivalent open-source contributions.

Culture & Benefits

• Competitive equity through stock options.
• More than 30 paid days off, including annual leave, UK bank holidays, and company closure days.
• Private healthcare with virtual and in-person care.
• Pension scheme with an 8% total contribution on full earnings.
• Daily breakfast, catered lunch, and snacks provided in the office.
• Direct collaboration with engineering, research, product, and founding leadership on hirify.global robotics.