Build the future of earth observation.

The Role

You own the perception stack that gives our drone eyes and understanding. From visual odometry and safety object detection to edge-ready scene understanding, you build and optimize the models that run on Jetson-class hardware and produce the autonomy signals the rest of the stack depends on.

What You'll Do

• Develop stereo visual-inertial odometry pipelines using NVIDIA cuVSLAM or similar approaches
• Build real-time 3D obstacle mapping for autonomous navigation
• Train and deploy safety detection models (people, vehicles, obstacles, landing zones)
• Deploy multimodal or vision-language models on edge hardware via TensorRT
• Support efficient edge-ready representations for constrained communications workflows
• Optimize multi-model inference to fit within Jetson-class memory, latency, and power budgets

Requirements

• Isaac ROS / ROS2 perception stack development
• NVIDIA TensorRT optimization: FP16/INT8 quantization, profiling
• Object detection: RT-DETR, YOLO-family, or similar edge-suitable architectures
• Edge deployment of multimodal or vision-language models on Jetson-class hardware
• Visual odometry (cuVSLAM, ORB-SLAM3, or similar)
• Camera calibration, synchronization, and perception-stack validation on real hardware
• ROS2 Humble: lifecycle nodes, QoS, multi-node systems
• Python + PyTorch for model development

Preferred

• NVIDIA Jetson Orin development experience
• nvblox real-time 3D reconstruction
• VQ-VAE or neural compression architectures
• TAO Toolkit for transfer learning
• Isaac Sim for synthetic training data
• Multi-model inference scheduling on shared GPU

The Role

You ensure every critical byte of data gets from a compact autonomous drone to the cloud through highly constrained connectivity paths. You own the RF design, modem behavior, antenna tradeoffs, and board-level signal isolation needed to make satellite and cellular links viable inside a compact flying wing airframe.

What You'll Do

• Integrate NTN and cellular modules for direct-to-cloud communication experiments
• Develop compact antenna systems within tight airframe constraints
• Design for RF performance in electrically noisy embedded environments
• Characterize RF interference from propulsion systems and high-speed digital electronics, then mitigate EMI
• Help define grounding, shielding, and mixed-signal layout practices for communications hardware
• Perform end-to-end link budget analysis for constrained connectivity paths
• Design practical low-bandwidth and store-and-forward communications behavior
• Optimize bandwidth utilization for TerraFirma's dual-token data architecture

Requirements

• Strong RF design background for compact embedded systems
• Antenna design for small platforms, impedance matching, radiation patterns, and RF measurement
• Experience dealing with noisy ground planes and isolating digital, analog, and RF sections
• Cellular and modem integration: AT commands, serial diagnostics, MQTT/CoAP, and embedded connectivity debugging
• Satellite communications: LEO NTN concepts, 3GPP NTN fundamentals, link budget analysis
• RF system integration inside compact UAV platforms
• Practical low-bandwidth networking and store-and-forward design
• Test equipment: VNA, spectrum analyzer, antenna pattern measurement
• Interference mitigation in constrained RF environments

Preferred

• Experience with NTN-capable cellular modules
• Antenna design for drones or small UAVs (<500g)
• LoRa mesh protocol development (beyond gateway mode)
• Direct-to-satellite IoT experience
• Sub-1W power budget communications systems
• 3GPP NTN Release 17 familiarity

The Role

You train the models that the AI/Perception Engineer architects. You build datasets, run training pipelines, manage experiments, and deliver INT8-quantized models ready for Jetson deployment. You help turn sparse early data and synthetic pipelines into deployable perception models that improve with each iteration.

What You'll Do

• Train safety detection models using aerial imagery datasets plus synthetic data augmentation
• Fine-tune multimodal and vision-language models for environmental scene understanding
• Develop edge-deployable model variants for constrained compute environments
• Implement confidence calibration for safety-critical predictions
• Manage experiment tracking, dataset versioning, and model validation pipelines
• Build export-ready training workflows for ONNX and TensorRT deployment

Requirements

• PyTorch model training at scale (multi-GPU, distributed)
• NVIDIA TAO Toolkit or equivalent transfer-learning workflow
• Dataset management: annotation pipelines, versioning, class balancing, and QA
• Model quantization: PTQ/QAT for INT8/FP16, TensorRT export
• Experiment tracking (MLflow, Weights & Biases, or similar)
• Azure GPU infrastructure management
• Understanding of precision/recall tradeoffs and calibration for safety-critical systems

Preferred

• Aerial/drone imagery training (VisDrone, DOTA datasets)
• NVIDIA Isaac Sim for synthetic data generation
• VQ-VAE or neural compression model training
• Confidence calibration techniques
• Embedded deployment targets (Jetson, edge TPU)
• Agricultural or environmental remote sensing domain knowledge

The Role

You own the electrical backbone of the prototype system, from custom board layout through bench integration. We need someone who can take reference designs and proven subsystems, adapt them into compact, manufacturable board layouts, and then help bring the full electronics stack together quickly with minimal hand assembly.

What You'll Do

• Design and refine compact PCB layouts for prototype avionics, carrier, interface, and power-distribution needs
• Adapt and integrate reference designs into manufacturable custom boards optimized for fast assembly and clean interconnects
• Build and validate bench-top electronics systems spanning compute, sensors, power, and communications modules
• Develop wiring and connector strategies that reduce solder-heavy assembly and speed prototype turnaround
• Validate sensor buses (I2C/SPI/UART) and troubleshoot signal integrity issues
• Debug EMI, grounding, and mixed-signal issues across compute, cameras, motors, and radios
• Build safe power-up, bring-up, and fault-isolation workflows for evolving prototypes
• Review, rework, and debug complex PCB assemblies when needed

Requirements

• Strong PCB design and layout experience in KiCad or similar EDA tools
• Experience adapting reference designs into custom board layouts for embedded systems
• Strong SMD soldering and rework skills, including fine-pitch components; IPC-A-610 Class 2+ preferred
• NVIDIA Jetson module integration or comparable embedded compute bring-up
• I2C/SPI/UART bus debugging with logic analyzer and oscilloscope
• Power systems integration: regulators, battery distribution, and power integrity
• Schematic reading, PCB debug, connector selection, and harness design for constrained systems
• Ability to bring up and debug complete bench-top avionics systems end to end
• Strong bench discipline around staged power-up, fault isolation, and prototype safety
• Comfort using AI-assisted engineering tools to accelerate layout and integration work

Preferred

• Experience with Jetson Orin family (NX or AGX)
• RF module integration (cellular/satellite/LoRa antennas)
• Sub-500g drone electronics experience
• Familiarity with open-source flight-controller or embedded control board ecosystems
• Familiarity with MicroXRCE-DDS (PX4 ↔ ROS2 bridge)