Senior Robotics SWE, System Identification & Modeling

About Nimble

Nimble is an AI robotics company building the autonomous supply chain to power fast, efficient and economical commerce. We’re training robot AGI to power a proprietary generalist supply chain superhumanoid, the first robot in the world capable of performing thousands of tasks across the supply chain. We’ve raised over $220 million at a valuation of over $1 billion and formed a strategic alliance with Fed Ex to build a national network of autonomous warehouses capable of generating many billions in annual revenue.

We are a hardcore, obsessed team of the world’s best engineers and operators. Our founding team comes from the AI labs at Stanford and Carnegie Mellon, and our board of directors includes robotics and AI legends such as Fei‑Fei Li, Marc Raibert, and Sebastian Thrun. We are on a mission to empower and inspire mankind to accomplish legendary feats by inventing robots that liberate us from the menial.

We are committed to building legendary products, a legendary team, and a legendary legacy. Join us and become part of an ambitious, humble, and resourceful culture where your work will leave a lasting impact on the future of robotics and commerce.

• Be relentlessly resourceful - Challenge conventions and overcome obstacles.
• Be legendary - Be the very best and do work that inspires.
• Be humble - Prioritize growth, learning, and doing whatever is needed to further the mission.
• Be dependable - Take ownership and deliver with high agency.

We are looking for a Senior Robotics Software Engineer specializing in System Identification and Modeling to build the core software powering our next‑generation autonomous robots. In this role you will develop and maintain mathematical models, parameter‑estimation pipelines, and feed forward/feedback control systems that allow our robots to operate with exceptional reliability, precision, and efficiency in real production environments. You will work across the full robotics and autonomy stack, building robust, production‑grade software that scales as we deploy more robots into high‑throughput operations, while serving as the team’s deep expert in system identification, dynamics modeling, and model‑based control.

You’ll collaborate closely with AI, hardware, controls, and infrastructure teams to integrate frontier AI capabilities with rigorous physics‑based models, continually improving robot uptime, performance, and overall intelligence.

• Design and execute system‑identification experiments for actuators, mechanisms, and full robot subsystems—motors, arms, elevators, drive trains—fitting dynamic models via regression and curve‑fitting to derive accurate feed forward and feedback controllers.
• Own kinematic calibration workflows: DH parameter identification, wheel radius estimation, and tool‑center‑point calibration to drive measurable improvements in arm accuracy and mobile base odometry.
• Build automated calibration and sys‑ that runs on production hardware, enabling rapid re‑characterization after mechanical changes, wear, or new platform deployments.
• Lead design and implementation of robot behaviors and task‑level intelligence across the full stack, integrating perception, planning, and control into reliable end‑to‑end execution across nominal and edge‑case scenarios.
• Drive measurable improvements in autonomy quality and arm accuracy using data, operational metrics, and model validation (diagnostic plots, residual analysis, statistical benchmarks).
• Collaborate with hardware engineering on software–hardware integration for new platforms and upgrades; triage and resolve production robotics issues.
• Lead technical design reviews, drive architecture decisions for core subsystems, and mentor engineers and technicians on reliability, testing, and operational excellence.

• Graduate‑level coursework or research in system identification, optimal estimation, or adaptive control.
• Experience with nonlinear system identification techniques such as extended Kalman filters, particle filters, or neural‑network‑based model learning.
• Familiarity with physics simulators (Mu Jo Co , Drake, Isaac Sim) and…