Autonomy Software, Behavioral Planning

ZipLine • South San Francisco, CA

61d • $200,000 - $265,000

About The Position

Zipline operates the world's largest autonomous flight network-and we're scaling from thousands to tens of thousands of carefully coordinated robots in dense, dynamic airspace. As a Motion Planning Software Engineer, you'll own the planning and decision-making software that turns perception and control into safe, efficient, and reliable flight at global scale. This is a rare IC seat where your algorithms leave the lab and run a fleet.

Requirements

• At least 8+ years building production software for safety-critical systems (aerospace/AV/robotics/medical).
• Strong in Rust/C++/C for real-time, fault-tolerant code on embedded/Linux.
• Depth in planning & search (A/ anytime / RRT/sampling), trajectory optimization/MPC, or multi-agent deconfliction-shipped on real robots/vehicles.
• Hands-on with simulation at scale, SIL/HIL, log replay, and metrics-driven validation.
• Evidence of shipping production-grade autonomy through ambiguous, noisy conditions-owning the last mile to reliability.
• Systems thinker who collaborates tightly with perception, controls, and flight ops; crisp docs and design reviews.

Responsibilities

• Build real-time trajectory generation and decision-making for autonomous flight (search-based, sampling, MPC, convex/non-convex optimization).
• Design collision avoidance and large-scale multi-agent planning (fleet deconfliction, airspace rules, traffic management) that scales to 10,000+ flights/day.
• Tackle joint optimization across safety, energy, time, and reliability-balancing mission goals with vehicle/airspace constraints.
• Plan in uncertain environments with complex dynamics: chance-constraints, robustness to wind/turbulence, degraded sensors, and partial observability.
• Extend the autonomy stack for new aircraft and payloads; define clean interfaces with perception and controls.
• Prove it before flight: scenario libraries, SIL/HIL, large-scale sim, log replay, and fault-injection.
• Mine real fleet data to validate safety metrics (separation, risk, intervention rate), improve models, and burn down long-tail failure modes.
• Continuously reduce cost/latency (vectorization, warm starts, smart heuristics) while raising reliability.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Industry: Professional, Scientific, and Technical Services
• Education Level: No Education Listed
• Number of Employees: 1,001-5,000 employees