Perception & Algorithms Engineer

Furientis • Los Angeles, CA

1d • Onsite

About The Position

Furientis is developing cost-effective, high-production-rate interceptor missiles to address a shortage of defensive munitions. The Seeker Team develops the weapon system’s sensing and targeting logic, including a low-cost, mass-producible, multimodal missile seeker. The team emphasizes technical ingenuity, rapid execution, and hardware that flies, in a broad-scope environment where engineers may take on work outside their specialty. It also uses an AI-centric workflow and expects engineers to contribute to how agentic coding tools are applied and governed.

The Perception & Algorithms Engineer will develop perception and estimation algorithms for the seeker, including detection, tracking, classification, and aimpoint localization using infrared imagery and fused sensor measurements. The role involves researching, implementing, optimizing, and validating algorithms, including AI and machine-learning methods where appropriate. Algorithms must meet strict latency requirements on embedded flight-compute hardware. The engineer will work with the Seeker Team to take algorithms from prototype through flight, validating performance in simulation, hardware-in-the-loop testing, and flight data.

The position is based on-site at Furientis’s Los Angeles headquarters; remote work is not available. Monthly weekend travel is required for test events and supplier engagements. A clearance is not required, but the candidate must be a U.S. Person and eligible to access export-controlled information. Furientis is an equal-opportunity employer. AI assists with application review, while humans make final hiring decisions.

Requirements

• Bachelor of Science in Optical Engineering, Computer Science, Electrical Engineering, Aerospace Engineering, Applied Math, Physics, or a related field.
• 5 to 7 years of experience in applied computer vision, perception, or estimation on real sensor data.
• Fluency in state estimation, including the Kalman filter family, nonlinear variants, and interacting multiple model approaches, applied with judgment.
• Experience with multi-target data association, such as JPDA, multiple hypothesis tracking, or modern equivalents.
• Working knowledge of classical optical image and signal processing, including spatial and temporal filtering, morphology, non-uniformity and fixed-pattern noise handling, and techniques for low-SNR imagery.
• Statistical rigor, including NEES and NIS consistency testing, covariance validation, and calibrated confidence.
• At least 3 years of experience using Python to create, alter, and maintain algorithm libraries within a modern Git and CI/CD framework.
• Must be a U.S. Person and eligible to access export-controlled information.
• Must work on-site at the Los Angeles headquarters; remote work is unavailable.
• Must be able to travel on weekends monthly for test events and supplier engagements.

Nice To Haves

• Master of Science or Ph.D.
• Rust experience for real-time implementation.
• Familiarity with embedded CUDA.
• Missile, interceptor, or guided-munition perception or tracking experience.
• Track-before-detect or other methods for detection and tracking at low SNR.
• IR or other non-visible-band sensor experience, including synthetic-to-real domain adaptation.
• TensorRT or comparable edge-inference deployment experience.
• Multi-sensor registration, boresight exploitation, or radar measurement familiarity.
• Modern machine-learning fluency, such as PyTorch or equivalent, including training, evaluating, and shrinking perception models with measured accuracy tradeoffs.
• Daily use of agentic coding tools such as Claude Code or similar.

Responsibilities

• Design and implement object-detection algorithms for MWIR and LWIR imagery that are robust to clutter, domain shift, and countermeasures.
• Develop multi-object tracking and state-estimation methods, including filtering, data association, and line-of-sight estimates for missile guidance.
• Implement track lifecycle management, including initialization, maintenance, confirmation, and termination under real sensor conditions.
• Develop segmentation, classification, and discrimination algorithms to separate targets from decoys, debris, and background.
• Develop aimpoint localization and supporting image-processing methods for terminal accuracy.
• Contribute to perception-related sensor fusion combining IR and X-band radar measurements into a consistent track picture.
• Implement, optimize, and validate algorithms for real-time flight, with measured latency, defensible covariance and confidence, and performance demonstrated in simulation, hardware-in-the-loop testing, and flight data.
• Research and prototype advanced AI and ML methods for perception and estimation, and mature successful methods into the flight pipeline.
• Contribute governance input and help build an AI-centric workflow using agentic coding tools.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level