We build humanoids that take the world’s most dangerous jobs so people don’t have to. Small team, real hardware, work that ships to the field.
Minerva Humanoids is developing rugged humanoid robots to do the most dangerous jobs on the planet, where human life is truly at risk. Our Robot Learning team is responsible for training policies that learn a variety of whole-body directable skills, from locomotion to manipulation and climbing.
As a full-time ML/AI Humanoid Robot Controls Engineer, you will be a core member of the Robot Learning team, responsible for bridging the gap between simulation and the physical world, ensuring our robots can operate across complex terrains and perform intricate tasks with reliability and precision. If it doesn’t work on the robot, it doesn’t work.
We believe that those working on life-saving missions should share in the success of that mission.
We build Western-sovereign humanoid robots for the world's most demanding environments. Backed by General Catalyst, with teams in San Francisco, London, and Germany.
Our engineers build the brains. You are an extension of the brains and also the hands. You take our humanoids out of the lab and put them to work in the field: desert heat, Arctic cold, onshore refineries, offshore steel. You own the robot from the moment the crate opens to the moment the job is done. When something breaks a thousand miles from HQ, you fix it. This role exists so our robots can be anywhere in the world, running, while the rest of the team keeps building.
We believe that those working on life-saving missions should share in the success of that mission.
Full time. Competitive salary and meaningful equity. Expect significant travel. Your office changes every week.
In person, five days a week.
Minerva Humanoids is developing rugged humanoid robots to do the most dangerous jobs on the planet, where human life is truly at risk.
Our robots combine high power, dense sensing, onboard compute, and high-bandwidth communications to tackle complex tasks in unstructured environments.
When you train a model to control a robot, it is difficult to answer: did this change make the robot better? How do we know?
You will build the software systems that help us answer those questions. Your first major project will be our policy evaluation platform, bringing training logs, simulation results, and real-robot recordings into one coherent system. It should make it easy to compare policies, inspect behavior through metrics, videos, and plots, and uncover recurring failures.
You will have ownership across the stack, including logging and reproducibility in training, simulation, and robot deployments. You will shape these systems and choose the right tools, with the chance to train and deploy policies yourself as you develop the platform. We make extensive use of AI agents, so these systems should be easy for both people and agents to use. The goal is to produce evidence that helps us decide what to train, test, or deploy next.
This role is open-ended, with substantial room for growth. You will work directly with machine-learning and robotics experts to identify bottlenecks and design and build systems that address them as the company scales.
We welcome early-career engineers and recent graduates who have demonstrated exceptional ability. We are not credentials-driven: we care more about competence, taste, curiosity, and rate of improvement than a particular degree. Show us what you have built.
We believe that those working on life-saving missions should share in the success of that mission.
Employer: Minerva Humanoids GmbH, Germany
Minerva Humanoids is developing rugged humanoid robots to do the most dangerous jobs on the planet, where human life is truly at risk. Our Robot Learning team is responsible for training policies that learn a variety of whole-body directable skills, from locomotion to manipulation and climbing.
As a full-time ML/AI Humanoid Robot Controls Engineer, you will be a core member of the Robot Learning team, responsible for bridging the gap between simulation and the physical world, ensuring our robots can operate across complex terrains and perform intricate tasks with reliability and precision. If it doesn’t work on the robot, it doesn’t work.
We believe that those working on life-saving missions should share in the success of that mission.
Employer: Minerva Humanoids GmbH, Germany
Minerva Humanoids is expanding its hardware team to develop the next generation of reliable and robust humanoid robots to handle some of the most dangerous jobs on the planet. From offshore rigs, refineries, and chemical plants to bomb disposal and hazardous substance decontamination, the work that requires sending humans into hazardous environments today will be handled by our robots tomorrow.
As the Head of Mechanical Hardware, you will work alongside other hardware leads to deliver future prototype iterations in line with Minerva’s long-term strategy - ultimately leading to a fully certified product. You will ensure that Minerva leverages the best available resources to design, build, certify, and manufacture. This will include growing the mechanical engineering team, engaging with technical partners and suppliers, and establishing in-house development and production facilities.
We believe that those working on life-saving missions should share in the success of that mission.
Employer: Minerva Humanoids GmbH, Germany
Minerva Humanoids is expanding its hardware team to develop the next generation of reliable and robust humanoid robots to handle some of the most dangerous jobs on the planet. From offshore rigs, refineries, and chemical plants to bomb disposal and hazardous substance decontamination, the work that requires sending humans into hazardous environments today will be handled by our robots tomorrow.
As a Senior Electrical Engineer, you will work alongside the hardware team to deliver prototype iterations in line with Minerva’s long-term strategy - ultimately leading to a fully certified product. You will be responsible for key areas of the robot electrical design, including architecture optimization, component selection, custom PCBs, harness and connectors, battery, and power distribution. As an early-stage hire, you will have the opportunity to advance rapidly alongside the company's growth.
We believe that those working on life-saving missions should share in the success of that mission.
Employer: Minerva Humanoids GmbH, Germany
Minerva Humanoids is expanding its hardware team to develop the next generation of reliable and robust humanoid robots to handle some of the most dangerous jobs on the planet. From offshore rigs, refineries, and chemical plants to bomb disposal and hazardous substance decontamination, the work that requires sending humans into hazardous environments today will be handled by our robots tomorrow.
As a Senior Mechanical Engineer, you will work alongside the hardware team to deliver prototype iterations in line with Minerva’s long-term strategy - ultimately leading to a fully certified product. You will be responsible for key areas of the robot mechanical design, including material selection, structural optimization, and the integration of actuators and transmissions. As an early-stage hire, you will have the opportunity to advance rapidly alongside the company's growth.
We believe that those working on life-saving missions should share in the success of that mission.
If you want to build Roger and none of the posted roles fit you, apply here for a full-time role or an internship.
Internships run 3 to 6 months. Please attach your CV.
Tell us about your experience and the work you want to do at Minerva.
We move fast. And we want you to move with us! See our list of open positions in the Minerva Talent Program below.
Our internships run 3 to 6 months. The longer you stay, the more real work you ship. And the more likely you can convince us to turn it into a full-time offer!
Ready to get started? Fill out the application form and we'll be in touch. If none of the positions below fit you, use the Open Application under Open roles.
Minerva Humanoids is developing rugged humanoid robots to do the most dangerous jobs on the planet, where human life is truly at risk. Our Systems team is responsible for the onboard computers, middleware, cameras, networks, and operator interfaces that carry commands, robot state, and live video between the robot and its operators.
You will help make the whole robot work as one system. Onboard computers, middleware, cameras, networks, and operator interfaces must carry commands, robot state, and live video reliably. You will own a focused project in teleoperation transport, middleware, robot bring-up, or diagnostics that improves one of those connections, and validate it on a real robot.
You will measure command and video latency, packet loss, and reconnection behavior for your project, improve installation, configuration, and startup checks so engineers can bring up a robot consistently, and debug hardware and software failures using logs, timing measurements, and packet captures. You will work with controls, perception, and deployment engineers to test stale commands, lost links, and recovery. Our robots operate in demanding environments where operators need responsive control and clear feedback, so your work should make setup repeatable, failures visible, and recovery predictable.
This role offers ownership of a substantial engineering project, with direct technical guidance from the founding team. We value initiative, organization, and persistence. We are flexible about the approach, provided that the solution works reliably on the robot.
Minerva Humanoids is developing rugged humanoid robots to do the most dangerous jobs on the planet, where human life is truly at risk. Our Robot Learning team is responsible for training policies that learn whole-body skills, from locomotion to manipulation and climbing.
You will develop locomotion and whole-body control policies for our humanoid robot: walking, stair climbing, terrain adaptation, and coordinated arm and leg motion in industrial environments. Policies are trained in simulation with reinforcement learning and deployed on the physical robot.
You will build on our existing training and sim-to-real pipeline, working on domain randomization, system identification, reward and curriculum design, and locomanipulation skills such as grabbing rails while climbing or opening hatches while balancing. You will benchmark locomotion primitives for speed, stability, and energy efficiency and evaluate policy performance on hardware.
This role offers ownership of a substantial engineering project, with direct technical guidance from the founding team. We value initiative, organization, and persistence. We are flexible about the approach, provided that the solution works reliably on the robot.
Minerva Humanoids is developing rugged humanoid robots to do the most dangerous jobs on the planet, where human life is truly at risk. Our Perception team is responsible for real-time vision, state estimation, and spatial understanding systems for locomotion, manipulation, and navigation.
You will develop perception and navigation capabilities for a humanoid robot with several onboard cameras: real-time localization and mapping, semantic understanding of industrial equipment, and autonomous navigation in GPS-denied facilities such as refineries, offshore platforms, and processing plants. Simulation will support development and large-scale evaluation before physical robot trials.
Build on our SLAM and navigation stack through one focused project. Mapping and navigation work may include persistent maps, relocalization, exploration, planning around measured locomotion limits, and recovery from blocked routes or lost localization. Industrial perception work may include equipment detection, instrument readings with uncertainty, automated annotation, 3D map and operator overlays, and accuracy and latency tests across varied conditions. Evaluate in simulation and validate on robot data.
This role offers ownership of a substantial engineering project, with direct technical guidance from the founding team. We value initiative, organization, and persistence. We are flexible about the approach, provided that the solution works reliably on the robot.