What are the responsibilities and job description for the Founding Robotics Software Engineer position at Kovari Industries?
About Kovari Industries
At Kovari, we're rethinking how physical work gets done in the age of robotics. We believe that robots capable of valuable, real-world labor are the new foundation for global prosperity.
Our company is developing general-purpose robots designed to take on the physical, repetitive work that keeps hotels running, starting with housekeeping and core hospitality workflows. By mastering this tractable domain, we're building the foundation for robots that will support an ever-expanding range of physical work.
Founded in San Francisco, Kovari is obsessed with rapid iteration, engineering rigor, and deploying real machines into real environments. The next decade will compress a century of progress in robotics, and we're looking for people who want to leave their fingerprints on that future.
About This Role
We're seeking a founding robotics software engineer to join our team and help build our robotics platform from the ground up. As a founding robotics software engineer, you'll work across the stack to address key technical risks and help implement our product roadmap.
What You'll Do
- Integrate vision–language–action (VLA) models into a custom robot with a unique drivetrain, elevator, and dual-arm system.
- Build adapter layers and policy wrappers that translate VLA outputs into safe, reliable robot control.
- Collaborate with controls engineers on IK, servo policies, and motion planning.
- Build end-to-end data pipelines (ROS 2 bags → synchronized observations/actions → tf-aligned training data).
- Design dataset curation, labeling, and augmentation workflows for fine-tuning.
- Fine-tune VLAs using behavior cloning, offline RL, and preference optimization.
- Develop simulation or synthetic data tools to accelerate training and cover edge cases.
- Define core metrics for success rate, latency, safety, and sim-to-real performance.
- Build tools for failure analysis (counterfactual replays, saliency inspection, etc.).
- Iterate quickly using real-world telemetry from deployed robots.
What You Bring
- Strong programming skills in C , Python, and ROS.
- Experience with motion planning, controls, and manipulation.
- Background in computer vision and ML for robotics.
- Familiarity with TF trees, SLAM, perception, and multi-sensor fusion.
Bonus Points
- Experience solving sim-to-real challenges for ML-driven robots.
- Ability to debug complex tf2 setups.
- Deep knowledge of ROS 2 conventions (frames, joint states, safety patterns).
- Built dataset pipelines (ros2 bag → tf-aligned tensors → model fine-tuning).
- Integrated custom sensors or written ros2_control drivers.
- Experience with SLAM, state estimation, or calibration.
- Knowledge of hardware interfaces (BLDC motors, CAN FD, IMUs, encoders, etc.).