Autonomy Engineer – Deep Learning

Autonomy Engineer – Deep Learning

Full-Time | Core Team

LAT Aerospace | Perception & Autonomy

At LAT Aerospace, we’re building India’s first clean-sheet hybrid-electric STOL aircraft and a next-generation autonomy stack powering aircraft for complex missions. Our team is kicking off a new chapter — developing the full flight-control, obstacle-avoidance, GNSS-denied navigation, and swarm-coordination layers that will define LAT’s autonomy architecture. As a Deep Learning Autonomy Engineer, you will build and deploy real-time neural models for object classification, scene understanding, semantic threat identification, and perception-assisted navigation.

What you’ll do?

• Design and train deep learning models for object classification, detection, semantic segmentation, threat identification, and multi-class scene understanding from aerial camera feeds.
• Apply LLMs and multimodal AI models for contextual reasoning on image data — assisting in early threat classification, prioritization, and mission-level decision support.
• Develop lightweight, real-time inference pipelines optimized for edge devices and constrained compute aboard UAVs.
• Build data-processing workflows for mining, curation, labeling, and augmentation from large volumes of real flight data.
• Use synthetic datasets, domain randomization, and simulation environments to accelerate training and improve model generalization.
• Integrate deep vision models with the autonomy pipeline: navigation, obstacle avoidance, swarm behavior, and mission control.
• Optimize models for latency, throughput, memory efficiency, and robustness under diverse lighting, motion, and environmental conditions.
• Create evaluation benchmarks and metrics to measure classification accuracy, threat-detection reliability, and real-time inference performance.

What we're looking for?

• Strong experience designing and training deep learning models for image/video tasks — classification, detection, segmentation, or tracking.
• Proficiency with PyTorch (preferred) or TensorFlow for model development.
• Ability to write clean, structured Python code; some C++ experience is a plus.
• Familiarity with CNNs, Vision Transformers, multimodal models, and modern architectures used in perception.
• Understanding of real-time deep learning constraints on embedded Linux systems.
• M.S. or Ph.D. in computer science, electrical engineering or related discipline

Why LAT?

• Core-team role building the autonomy brain of India’s most ambitious aerospace startup.
• Ownership, speed, and the opportunity to define a new class of systems from scratch.