Software Engineer - Behaviors & Planning

Overland AI•Seattle, WA

3d•$160,000 - $190,000•Onsite

About The Position

Overland AI is seeking a Software Engineer to join the team developing the behaviors and planning stack that determines where autonomous ground vehicles go and how they get there in off-road terrain, including environments with degraded GPS and intermittent communications. The engineer will work alongside planning, optimization, and systems engineers, ship code to vehicles early, and grow toward co-owning a core part of the stack within the first 6–12 months. The role emphasizes strong fundamentals, curiosity, and end-to-end ownership rather than extensive production experience. The initial focus area will be matched to the candidate’s interests and team needs: systems architecture, performance, tooling, and reliability; planning and decision-making under uncertainty using learned and classical approaches; or trajectory optimization, including MPC/MPPI, vehicle modeling, and real-time optimization. The position is onsite in Seattle, Washington, and includes periodic field testing. The listed annual pay range is $160,000–$190,000. Actual level and base salary are determined case by case based on job-related knowledge and skills, education, and experience. Benefits include equity compensation, healthcare, dental and vision plans, flexible PTO, a 401(k) with company match, paid parental leave, a monthly wellness stipend, and a one-time workplace and productivity reimbursement. Employment is contingent on the ability to comply with U.S. export control laws because the position may involve access to export-controlled technology.

Requirements

BS or MS in Computer Science, Robotics, Engineering, or a related field, or equivalent experience; PhD graduates are welcome.
1–3 years of professional experience; internships, research, and substantial projects count.
Solid programming aptitude in C++ or Python and a desire to become excellent in both.
Strong computer science fundamentals, including data structures, algorithms, and complexity.
Enthusiasm for supporting periodic field testing.
Ability to comply with U.S. export control laws.

Nice To Haves

PhD in a relevant field.
Experience with both learned planning approaches, such as learned costmaps, imitation learning, or reinforcement learning, and classical approaches, such as search, sampling, or optimization.
Experience integrating prior or opportunistic map data with onboard perception.
Off-road UGV and/or on-road autonomous vehicle planning experience; UGV experience is a plus, not a requirement.
Experience in DDIL environments.

Responsibilities

Implement, test, and improve features in the behaviors and planning stack using C++ and Python in a shared codebase.
Use off-road driving data to evaluate changes and identify regressions.
Debug issues using logs, simulation, and field data, with mentorship.
Participate in field tests and observe code running on vehicles.
Participate in design discussions and code reviews, giving and receiving feedback.