Overland AI
Director of Software Engineering - Behaviors and Planning
$200K–$300K
Seattle, WA · On-site · Director · Full-time · Posted Jul 1, 2026
- Experience
- 5+ years beyond the PhD; 2+ years leading an engineering team.
- Education
- PhD in robotics, computer science, controls, or a closely related field, with a research focus in motion planning, autonomous navigation, or robot decision-making.
- Visa sponsorship
- Must be a U.S. Person, as defined under ITAR, and eligible to obtain and maintain a U.S. security clearance. Active Secret clearance is preferred but not required at the time of hire.
About the job
Leads the Behaviors & Planning team building and verifying single-vehicle autonomy for ground vehicles operating across road and off-road environments.
What you'll do
- Manage and mentor 7–10 engineers, set technical priorities, and shape the team’s roadmap and architecture.
- Lead planning, behaviors, geospatial navigation, and offline verification, including simulation, log playback, and regression testing.
- Partner with ML & Perception, support field deployments and demonstrations, and integrate emerging autonomy capabilities.
What they ask for
- PhD in robotics, computer science, controls, or a closely related field, with a research focus in motion planning, autonomous navigation, or robot decision-making.
- 5+ years of professional industry experience beyond the PhD shipping autonomy software.
- 2+ years leading an engineering team, including setting technical direction and delivering complex systems.
- Deep expertise in at least two listed areas; experience shipping autonomy software to hardware in unstructured environments.
Nice to have
- Off-road or unstructured-terrain navigation; ROS2 or similar robotics middleware; machine learning for planning or perception-planning integration.
- Geospatial planning or GIS for autonomy; defense or government contracting; simulation and log-replay verification systems.
Pay
$200,000 - $300,000 annually (base salary)
Tools
MPPI, RRT, ROS2, costmap, simulation, log playback, GIS, VLM
AI focus
Integrates terrain-understanding outputs from ML and Perception with classical planning, advances toward end-to-end learned planning, and explores VLM-based reasoning for behavioral guardrails, scene understanding, and contextual decisions.
Model-written summary · the original posting is authoritative