Senior Software Engineer, Task Group (R3014)

Introduction to Shield AIFounded in 2015, Shield AI is a venture-backed defense technology company whose mission is to protect service members and civilians with intelligent systems. In pursuit of this mission, Shield AI is building the world’s best AI pilot. Its AI pilot, Hivemind, has flown a fighter jet (F-16), a vertical takeoff and landing drone (V-BAT), and a quadcopter (Nova). The company has offices in San Diego, Dallas, Washington DC and abroad. Shield AI’s products and people are currently in the field actively supporting operations with the U.S. Department of Defense and U.S. allies.

The Planning and Controls (PAC) team is an agile group of engineers focused on researching and developing state-of-the-art algorithms that drive intelligent and confidence-inspiring flight behaviors while accounting for an uncertain and dynamic world.

As a member of PAC, you will work at the intersection of artificial intelligence, task & motion planning, and controls. You will architect core systems, write high quality software for core systems, set standards for software engineering, refine technical requirements, drive strategic technical improvements, and mentor other engineers.

What you'll do:

• Research, design, and implement state-of-the-art planning algorithms, taking ideas from conception to field deployment
• Optimize the planning stack on the CPU and GPU
• Develop core systems that must be portable and scalable across different UAS platforms and compute architectures
• Work with our Hivemind Design Software team to design and build infrastructure for the PAC team that both supports local development and scales into the cloud
• Work with soft or hard real-time operating systems
• Work with robotics middleware such as ROS and PX4
• Work on algorithms, including graph search, sampling-based planning, etc.
• Work with and deploy our AI stack to edge devices
• Work in collaboration with our DevOps teams to architect and maintain our cloud infrastructure, working with containerization technologies such as Docker and Kubernetes; assist with cross-platform builds and test orchestration, working with build tools (e.g. CMake, Bazel, etc.) and compiler tool chains (e.g. gcc, llvm, etc.)

Projects you might work on:

• Design and implement Mission Planning interfaces or Multi-agent coordination for Tasks given to heterogeneous teams.
• Port PAC libraries, where applicable, to CUDA kernels
• Port PAC libraries to EdgeOS, a Shield AI Middleware framework, that enables portability of PAC libraries to multiple platforms and compute architectures
• Design and implement an analysis and simulation framework that: enables testing of various components in isolation or in a fully integrated manner; supports parameter sweeps; supports hardware-in-the-loop

Required qualifications:

• Typically requires a minimum of 5 years of related experience with a Bachelor’s degree; or 4 years and a Master’s degree; or 2 years with a PhD; or equivalent work experience.
• Proficient with C++ 11, 14, or 17 in a production environment (e.g. support for both internal and external customers)
• Experience in a collaborative development environment, focused on continuous integration/delivery & test-driven development
• Strong knowledge of modern software engineering best practices
• Demonstrated record of working hard, strong communication skills, being a trustworthy teammate, and being kind to others
• Familiarity with writing thread-safe multi-threaded programs
• Familiarity with common debugging and profiling tools (e.g. gdb, valgrind, etc.)
• Familiarity with git, bash, Python