AI and Computer Vision Engineer
The Exploration Company5 hours ago
Munich, GermanyMid Level
Responsibilities
- Design, train, and evaluate deep-learning models for 6-DoF spacecraft pose estimation.
- Own dataset generation, management, augmentation, domain adaptation, experiment tracking, and reproducibility across synthetic, laboratory, and orbital imagery.
- Optimize models using knowledge distillation, pruning, quantization, and quantization-aware training for flight-representative compute.
- Port and evaluate models on embedded and neuromorphic hardware while measuring accuracy, latency, memory, and power.
- Build explainability and uncertainty capabilities and debug failure modes such as high occlusion.
- Prototype privacy-preserving, distributed, and self-supervised training approaches for model improvement and in-flight adaptation.
- Define requirements, test scenarios, and validation criteria with GNC/FPO teams and support space-qualified camera sensor selection.
- Run hardware-in-the-loop validation campaigns and analyze results.
- Manage training workloads across cloud GPU resources and internal HPC systems.
Requirements
- MSc or PhD in computer science, electrical engineering, robotics, aerospace, physics, or a comparable field with a strong machine learning focus.
- At least 3 years of experience building and shipping deep-learning computer vision systems; PhD work in the field counts.
- Demonstrable experience taking models from research prototypes to constrained targets through quantization, distillation, latency optimization, or embedded and accelerator deployment.
- Experience training on synthetic data and addressing the sim-to-real gap.
- Hands-on experience with cameras, calibration, laboratory test setups, and collecting and annotating data.
- Strong Python and PyTorch skills, or experience with JAX or TensorFlow.
- Knowledge of classical computer vision and 3D geometry, including camera models, intrinsics and extrinsics, distortion, PnP, RANSAC, and coordinate frames.
- Comfort with Linux, Git, containers, and large training jobs on GPU clusters or in the cloud.
- Ability to work independently and communicate results in technical reports and reviews.
- Working proficiency in English; German is a plus.
- Experience with spacecraft rendezvous, docking, vision-based navigation, SPEED/SPEED+, or ESA pose-estimation challenges is desirable.
- Familiarity with spiking neural networks, neuromorphic hardware, event-based cameras, federated learning, differential privacy, secure aggregation, synthetic-data rendering, or space and safety-critical software assurance is desirable.
Benefits
- Relocation assistance is provided for candidates willing to relocate, including visa sponsorship where applicable.
- Employment offers are contingent upon a background check, with processing handled through Zinc Work Limited when applicable.
- The company offers an open, collaborative, and fast-moving work environment focused on advancing space transportation.
Categories
About The Exploration Company
The Exploration Company designs and builds Nyx, a reusable cargo spacecraft for in‑orbit transportation and Earth return, selling mission services to commercial and institutional customers. Founded in 2021 and headquartered in Gauting, Bavaria, the privately held company focuses on logistics to and from low Earth orbit and beyond. It flew a reentry technology demonstrator on Ariane 6’s inaugural launch in 2024 as a step toward operational missions.