Harmattan AI · Defence and security
Machine Learning Research Engineer – Visual Navigation
- Zurich (On‑site), +2 more: Lausanne, Paris
- Full-time
AI roleMachine LearningTrains models
Still open on Harmattan AI's site · checked 4 hours ago
- Education
- A degree
- Posted
What you'll do
As written by Harmattan AI · Read the original
Responsibilities
DL Algorithm Development: Design deep learning models to analyze a drone's environment in multiple modalities and enable its localization through time and space.
Edge AI Optimization: Embed models on heterogeneous platforms with limited shared compute. Optimize trade-offs between model speed and accuracy to minimize pipeline latency.
Cross-Team Collaboration: Partner with the VIO, Navigation, and Embedded Software teams to build a robust, real-time localization pipeline, communicating trade-offs clearly to stakeholders.
Research & Innovation: Stay at the absolute forefront of scientific trends in Deep Learning to introduce cutting-edge methodologies to the team.
Pipeline Management & MLOps: Build, heavily modify, and manage training, evaluation, and MLOps pipelines while ensuring reproducibility, robust logging, and version control.
Validation & Field Testing: Actively participate in real-world flight tests. Analyze in-field failures to continuously iterate and improve deployed models.
Requirements
Candidate Requirements
Educational Background: Strong academic record with a STEM degree (e.g., Computer Science, Machine Learning, Robotics, Mathematics).
Deep Learning & Computer Vision Expertise: Proven track record of researching and productizing Deep Learning models for Computer Vision in the industry (Drones, Robotics, Autonomous Vehicles, etc.)
Technical Competence: Proficiency in Python and DL frameworks (PyTorch/Lightning). Experience with MLOps and edge deployment optimization.
Skills
Stack
- Python
- PyTorch
Working languages
English
Company
Last checked . The details above were extracted automatically; the employer's original posting is the authority.
