Perception & Fusion Engineer
- Role
- AI / ML
- Employment
- Full-time
- Company size
- Startup
Open to US only. Set where you work from to check your eligibility.
No BS summary
Engineer with MS/PhD or equivalent experience in robotics/CS/EE/applied math, specializing in perception and sensor fusion for UAVs. Requires expertise in estimation theory, multi-target tracking, deep learning, and strong programming skills in C++, Rust, Go, and/or Python.
Core skills
Required skills
Optional skills
Swarm Aero is redefining air power, building the largest swarming UAV and the most versatile swarming aircraft network in the world. The company is moving quickly to launch the first aircraft designed specifically for swarming, as well as the Command & Control software to mobilize swarms of thousands of heterogeneous autonomous assets and empower human operators to achieve superhuman results.
The team has created and exited multiple startups, negotiated defense deals worth billions of dollars, and designed and built 30+ novel aircraft, with aerospace experience from Scaled Composites, Airbus, Archer Aviation, Blue Origin, and Boom Supersonic.
About the Role
We're looking for a Perception & Sensor Fusion Engineer to develop the algorithms that give the largest drone swarm in the world an accurate, shared picture of the battlespace. This is a deep algorithm role for someone who lives in the details of detection, tracking, and estimation, and wants to see their work fly.
What You'll Do
- Train, tune, and test automatic target recognition and track management systems using the latest advancements in neural networks.
- Develop multi-sensor fusion and state estimation algorithms (Kalman filters, particle filters, multi-target tracking) running onboard resource-constrained vehicles.
- Design distributed fusion approaches that combine tracks across the swarm into a single coherent picture.
- Rigorously characterize algorithm performance against real-world flight data and simulation.
- Optimize models and estimators for real-time inference on edge compute.
- Write clean, maintainable, and efficient code.
- Travel up to 25% of the time for onsite test and integration events.
Basic Qualifications
- MS or PhD in Robotics, Computer Science, Electrical Engineering, Applied Math, or related field, or equivalent depth of applied experience.
- Demonstrated expertise in estimation theory and multi-target tracking (Kalman/particle filters, JPDA, MHT, random finite sets, or similar).
- Hands-on deep learning experience (training, evaluation, and deployment of neural networks for detection or classification).
- Strong programming skills in C++, Rust, Go, and/or Python.
- Track record of taking algorithms from research to deployment on a real system.
Preferred Qualifications
- Publications or deployed work in ATR, multi-sensor fusion, or distributed estimation.
- Experience with EO/IR, radar, or RF sensor processing on aerial platforms.
- Experience deploying ML models to embedded or edge hardware (TensorRT, ONNX, quantization).
- Familiarity with defense sensing and tracking problem domains.
What We Offer
- Meaningful equity in a high-growth defense technology company
- Competitive base salary commensurate with experience
- Medical, dental, vision, and 401k
- PTO and paid sick leave
- Monthly wellness stipend
- Daily catered lunch (office)
- Paid parental leave
- Direct impact on a critical national security mission
- A world-class team of engineers and operators solving genuinely hard problems
What you'll do
- Train, tune, and test automatic target recognition and track management systems using the latest advancements in neural networks.
- Develop multi-sensor fusion and state estimation algorithms (Kalman filters, particle filters, multi-target tracking) running onboard resource-constrained vehicles.
- Design distributed fusion approaches that combine tracks across the swarm into a single coherent picture.
- Rigorously characterize algorithm performance against real-world flight data and simulation.
- Optimize models and estimators for real-time inference on edge compute.
- Write clean, maintainable, and efficient code.
- Travel up to 25% of the time for onsite test and integration events.
What they require
- MS or PhD in Robotics, Computer Science, Electrical Engineering, Applied Math, or related field, or equivalent depth of applied experience.
- Demonstrated expertise in estimation theory and multi-target tracking (Kalman/particle filters, JPDA, MHT, random finite sets, or similar).
- Hands-on deep learning experience (training, evaluation, and deployment of neural networks for detection or classification).
- Strong programming skills in C++, Rust, Go, and/or Python.
- Track record of taking algorithms from research to deployment on a real system.
- Preferred: Publications or deployed work in ATR, multi-sensor fusion, or distributed estimation.
- Preferred: Experience with EO/IR, radar, or RF sensor processing on aerial platforms.
- Preferred: Experience deploying ML models to embedded or edge hardware (TensorRT, ONNX, quantization).
- Preferred: Familiarity with defense sensing and tracking problem domains.
Benefits
- Meaningful equity in a high-growth defense technology company
- Competitive base salary commensurate with experience
- Medical, dental, vision, and 401k
- PTO and paid sick leave
- Monthly wellness stipend
- Daily catered lunch (office)
- Paid parental leave
- Direct impact on a critical national security mission
- A world-class team of engineers and operators solving genuinely hard problems
Swarm Aero is redefining air power, building the largest swarming UAV and the most versatile swarming aircraft network in the world. The company is building aircraft designed specifically for swarming and Command & Control software for thousands of heterogeneous autonomous assets.