The posting
Field AI is transforming how robots interact with the real world. We are building risk-aware, reliable, and field-ready AI systems that address the most complex challenges in robotics, unlocking the full potential of embodied intelligence. We go beyond typical data-driven approaches or pure transformer-based architectures, and are charting a new course, with already-globally-deployed solutions delivering real-world results and rapidly improving models through real-field applications.
Learn more at https://fieldai.com.
What You'll Get to Do
1. Own End-to-End Robot Verification & Validation
- Design and execute verification and validation strategies for complete robotic systems, from individual capabilities through full autonomous missions
- Develop clear acceptance criteria, performance metrics, and test methodologies for new robot capabilities
- Validate system behavior across multiple robotic platforms, environments, and operating conditions
- Build repeatable qualification and regression processes that allow new capabilities to ship without compromising existing functionality
- Establish a clear understanding of what “deployment-ready” means and provide quantitative evidence that systems meet that bar.
2. Build Scalable Robotics Test Infrastructure
- Develop automated test infrastructure spanning simulation, hardware-in-the-loop, lab testing, and full robot operation
- Create reusable test scenarios and evaluation frameworks that exercise autonomy under nominal, edge-case, and failure conditions
- Build tools for experiment execution, telemetry collection, automated analysis, visualization, and reporting
- Improve the reproducibility of robot testing so failures can be recreated, diagnosed, and verified efficiently
- Help move validation from individual one-off tests toward continuously running, scalable system evaluation
3. Validate Real-World Robot Behavior
- Design tests that expose robots to the uncertainty and variability encountered in real deployments
- Exercise systems across changing terrain, obstacles, environmental conditions, sensor degradation, communication failures, compute limitations, and other realistic disturbances
- Evaluate not only whether a robot succeeds, but how reliably, safely, and consistently it behaves across repeated trials
- Identify performance boundaries and characterize where system behavior begins to degrade
- Work directly with physical robots in the lab and field to reproduce difficult system-level failures
4. Turn Failures Into Engineering Signal
- Debug failures across autonomy, sensing, state estimation, planning, control, system integration, compute, networking, and hardware boundaries
- Use telemetry and experimental data to isolate root causes rather than simply identify symptoms
- Develop tooling and instrumentation that make complex robot behavior easier to understand
- Convert field failures and difficult-to-reproduce issues into deterministic regression tests whenever possible
- Partner with subsystem owners to verify fixes and prevent recurrence
5. Drive System Reliability and Release Readiness
- Partner closely with autonomy, robotics software, hardware, systems, and field teams throughout the development lifecycle
- Identify integration and reliability risks early and ensure they are represented in the validation process
- Build dashboards, scorecards, and automated evaluations that provide a clear view of system health and capability maturity
- Help define release gates based on measurable system performance rather than subjective readiness
- Continuously improve the V&V process as the autonomy stack, robot platforms, and deployment environments evolve
What You Have
- Bachelor’s, Master’s, or PhD degree in Robotics, Computer Science, Electrical Engineering, Mechanical Engineering, or a related technical field, or equivalent hands-on industry experience
- Strong software engineering fundamentals with proficiency in C, C++, and Python for developing, integrating, debugging, and testing robotic systems
- Hands-on experience developing, integrating, testing, or debugging physical robotic or autonomous systems
- Broad understanding of robotic systems, including sensing, state estimation, planning, control, communication, compute, and actuation
- Experience developing and debugging robotics software using ROS 2 and distributed message-passing architectures
- Strong experience working in Linux environments, with familiarity with real-time systems and RT-patched Linux kernels
- Proficiency with Git and modern collaborative software development practices, including code review, branching, integration, and regression workflows
- Experience diagnosing complex system-level failures where root causes may span software, middleware, networking, compute, sensors, and hardware
- Experience designing quantitative experiments, defining performance metrics, and analyzing robot telemetry and test data to evaluate system behavior and reliability
- Strong instincts for testability, observability, reproducibility, and systematic debugging, with the ability to turn ambiguous robot behavior into concrete hypotheses, experiments, and actionable engineering work
- Experience developing system-level V&V, qualification, or release processes for autonomous robots, vehicles, drones, or other complex physical systems
- Experience building automated regression testing or large-scale evaluation infrastructure for robotics.
- Familiarity with simulation, hardware-in-the-loop, fault injection, or scenario-based testing
- Experience defining reliability metrics, operational performance envelopes, or quantitative release criteria
- Experience with automated telemetry analysis, experiment management, or fleet-level performance evaluation
- Experience testing robotic systems in unstructured, dynamic, or safety-critical environments
- Experience bringing new robotic capabilities from initial integration through production deployment
- A track record of finding subtle system-level problems that were difficult to reproduce or crossed traditional subsystem boundaries
- The ability and willingness to get hands-on with robots, understand the complete system, and go wherever the problem leads



