Senior ML Engineer (f/m/x) - Computer Vision for Earth Observation

Mission of this position (overall summary of the purpose, some comments about our current and future projects):

We are looking for a Senior ML Engineer (f/m/x) to build and scale computer vision systems for Earth observation across a range of problems: detection, segmentation, classification, change analysis, and geometric vision such as 3D reconstruction and image matching. You'll work primarily with very high resolution optical imagery and, increasingly, across modalities including synthetic aperture radar (SAR). You'll turn this data into reliable, production-grade capabilities that downstream products depend on, working across the full ML lifecycle from data and training through evaluation to deployment.

An important and growing part of this role is bringing models to where they run under tight constraints. LiveEO is building Twinspector, its own very high resolution satellite constellation, and running vision models efficiently under limited compute, memory, and power (including on-device and toward onboard processing) is increasingly central to what we do. You'll help take models from the training cluster to constrained environments through techniques such as model compression, quantization, and optimization.

This is a balanced role: part applied research, part engineering, all impact. The exact balance depends on your strengths, and we are open to profiles that lean more toward applied research or more toward engineering as long as the fundamentals are strong.

LiveEO is a young, dynamic team that thrives on big challenges and fast learning cycles. We move quickly, stay curious, and genuinely enjoy building together. We're on a mission to break the "curse of Earth Observation": turning incredible satellite data into reliable, actionable decisions that people can trust and use in real operations.

Cross functional collaboration:

You'll be part of Sektion 4, LiveEO's government-solutions product team. Sektion 4 owns its roadmap and delivers funded R&D projects end-to-end, from research through to production, and sets its own technical direction. You'll collaborate with other LiveEO teams and with external research partners while retaining ownership of the team's goals and deliverables. As the role extends toward on-device and onboard inference, you'll also work with the teams building LiveEO's Twinspector satellite programme. You'll work closely with our data and annotation function to define labeling and quality guidelines and to close feedback loops on data quality across geographies and acquisition conditions.

Tech stack & tools, which potential candidate will work with:

• Core ML: Python, PyTorch + PyTorch Lightning

• Experimentation: MLflow (tracking, model registry)

• Compute & orchestration: Ray (distributed compute), Prefect (workflows)

• Infrastructure: AWS and secure on-prem environments

• Edge / on-device: model compression and optimization for constrained and embedded targets

• Geospatial: GDAL, Rasterio, GeoPandas, STAC

• Datastores: PostgreSQL (metadata / operational data)

As a Senior ML Engineer, you will develop state-of-the-art computer vision systems that extract reliable information from large volumes of satellite imagery, across a range of tasks and all the way to production.

• Build vision models across tasks: design, train, and iterate on models for detection, segmentation, classification, change analysis, and geometric vision (3D reconstruction, image matching/registration) on VHR satellite imagery, with clear ablations and measurable performance improvements.

• Research to production: identify and adapt state-of-the-art approaches across computer vision and remote sensing (papers → prototypes → validated baselines), focusing on pragmatic wins under real constraints.

• Push toward the edge: bring models to constrained and embedded targets for LiveEO's Twinspector programme, using model compression, quantization, and optimization for on-device and onboard inference.

• Build scalable pipelines: training and evaluation infrastructure across cloud and secure on-prem environments, with experiment tracking, reproducibility, and systematic failure analysis across geographies and acquisition conditions.

• Deliver production-ready components: robust inference interfaces, model packaging, deterministic evaluation, and monitoring.

• Collaborate and communicate: work with the data annotation function on labeling guidelines and edge cases, with partner teams to turn model capabilities into validated deliverables, and with external researchers, presenting findings clearly and efficiently.

• Strong computer vision and machine learning fundamentals (representation learning, supervision strategies, evaluation design) and practical debugging/optimization skills.

• Practical experience across multiple computer vision problem types (for example detection, segmentation, classification, change detection, or geometric vision such as 3D reconstruction and image matching), with real depth in at least one. This can come from remote sensing or from another imaging domain where these problems matter (robotics, autonomous driving, photogrammetry, AR/VR, medical imaging).

• Strong Python engineering fundamentals with clean, maintainable code, and deep experience with PyTorch, implementing and training deep learning models at scale.

• Strong understanding of ML experimentation, versioning, and tracking.

• Background in remote sensing, computer science, physics, or a related field, or equivalent practical experience.

• Comfortable working with researchers and presenting findings clearly and efficiently.

• Willingness to work on-site in Berlin and eligibility to obtain a German security clearance (Sicherheitsüberprüfung).

Nice to have:

• Experience deploying models under constrained compute or on edge / embedded devices (model compression, quantization, optimization).

• Hands-on experience with satellite / remote-sensing imagery (strongly preferred).

• PhD in remote sensing, computer science, physics, applied mathematics, or a related field.

• Distributed computing with Ray; workflow orchestration with Prefect (or similar).

• Cloud platforms (AWS) and/or secure on-prem / HPC experience (SLURM, Docker, DVC).

• Geospatial stack: GDAL, Rasterio, GeoPandas, STAC.

• PostgreSQL (or similar).

• Experience with SAR alongside optical imagery, or familiarity with geospatial foundation models / VLMs (self-supervised, contrastive, masked modeling).

• Experience with post training VLMs (SFT, RLVR).

• Experience with synthetic data generation and sim2real / domain adaptation.

• The opportunity to create a product that can improve business processes and lives across the globe.

• Flexible working hours and hybrid work model - we trust our employees to get their work done while maintaining a healthy work-life balance.

• We empower employees to drive their own career development, take initiative and have the freedom to be creative and bold.

• Not an overtime culture - we take care that overtime is done only as a necessity and always offset with time off and rest.

• A collaborative and learning environment - frequent internal workshops, knowledge sharing sessions, journal clubs and hackathons.

• Office located in the centre of Berlin Kreuzberg with free fruit, nuts and drinks.

• Potential to participate in the employee stock option program.

• Urban Sports membership and BVG subsidy, corporate pension program.

• A diverse and vibrant international environment of 30+ different nationalities