Founding Robotics Engineer, Perception & VLA

Destro; New York (NY); On-site; USD 150,000 - 230,000; Full time

Benefits offered by this job

Founding equity

Job summary

Destro is hiring a robotics perception engineer to own the perception and learning stack for VisionOS, taking models from simulation to on‑floor production and helping robots understand and act in real warehouses.

You’ll own the full model lifecycle, deploy on real robots with tight latency, work with a founding team, and receive founding equity as the company scales; this role involves travel to customer sites.

Qualifications

• 3–5 years in robotics, CV, or ML with physical robots.
• Hands-on GPU model training and production inference under latency limits.
• 3D perception with point clouds and depth cameras.
• Experience deploying on real robot hardware and in simulation (MuJoCo / Isaac Sim).

Responsibilities

• Own perception and learning stack from model development to deployment.
• Develop VLA models and baseline pre-training/fine-tuning.
• Ensure models run on real robots with low latency.
• Plan perception/learning roadmap with the founding team.

Skills

Robotics

Computer vision

Machine learning

Python

C++

Real hardware deployment

Simulation environments

Tools

MuJoCo

Isaac Sim

Job description

Walk into any busy distribution center and you'll see the problem we're solving. Thousands of products, packaging that changes every season, light that shifts every time a dock door opens. Most picking and case handling is still done by hand, because robots haven't been able to make sense of a scene like that and act on it reliably.

We're hiring the engineer who will lead how our robots see and learn. You'll own perception and robot learning for VisionOS, make the core technical calls, and take our models from simulation onto real robots on customer floors.

Destro is building the intelligence layer for robotics. MothershipOS coordinates robots, people, and workflows across an entire operation in real time, and it's already running inside large enterprise operations. VisionOS brings that intelligence to the individual robot: perceiving what's in front of it, deciding what to do, and handling it. Our founders previously led teams at HAI Robotics, Dexterity, and Matic, and we closed our Seed round at one of the highest valuations for a robotics company at this stage.

About the Role

This is a founding role, and you'll build this part of the product from the ground up with the founding team. There isn't a large existing codebase waiting for you. The perception team recently lost a member, and we need someone who wants to build the next version from scratch, set the technical approach, and operate with a high degree of independence.

Your work runs through the full model lifecycle: perception pipelines, VLA models, pre-training, fine-tuning, deployment, and post-training optimization all the way to inference. For us, a model is done when it works on the floor, not when the eval curve looks good.

Why You'll Want This Job

• You set the direction. You'll own the perception and learning stack and decide how we improve and build it.
• Your work goes live fast. Our customers' warehouses are our lab, so your models get tested in real operations in months, not years.
• Startup speed. Small team, fast decisions. If you can make the case for something, you can start building it that week.
• You grow as we grow. Your scope expands with every new task, robot, and customer, alongside founders who have scaled robotics before.
• Founding equity and a real stake in where the company goes.

What You'll Own

• Perception: 3D scene understanding (RGB-D fusion, point clouds, segmentation), 6-DOF pose estimation for products and cases, and zero-shot identification with vision-language models.
• VLA models: Vision-language-action models as the backbone of how our robots carry out tasks, including pre-training, fine-tuning, and synthetic data in simulation.
• Deployment: Models running on real robots within tight latency and compute limits, plus an active learning loop that turns floor mistakes into better models.
• Data and teleoperation: Teleoperation setup, data collection, and the pipeline from raw recordings to clean, versioned training data.

Your First Month

Get to know our robots, simulation environment, and customer floors, then write the perception and learning plan you want to bet on. By the end of the month, you'll have a first model running on a physical robot doing a real task.

What You Bring

• 3 to 5 years of hands-on work in robotics, computer vision, or ML, ideally on physical robots
• Hands-on model training on GPUs and experience shipping models to production inference under latency constraints
• 3D perception experience with point clouds and depth cameras
• Experience deploying on real robot hardware and working in simulation (MuJoCo, Isaac Sim, or similar)
• Strong Python and C++ (or a similar systems language)
• A habit of getting physical prototypes working in weeks, not months

Nice to Have

• VLA or manipulation policy training
• Grasp planning, learned or geometric
• Warehouse or logistics automation experience

Travel

This role is travel-heavy. Our customer warehouses are our lab, so expect to spend a significant amount of your time on site, deploying and improving systems on live floors. If you need to be home most weeks, this probably isn't the right fit, and we'd rather say so now.