Senior AI Engineer Space42

Overview

ABOUT US

Space42 (ADX: SPACE42) is a UAE-based AI-powered SpaceTech company that integrates satellite communications, geospatial analytics and artificial intelligence capabilities to enlighten the Earth from space. Established in 2024 following the successful merger between Bayanat and Yahsat, Space42’s global reach allows it to address the rapidly evolving needs of its customers in governments, enterprises, and communities.

Our vision is to pioneer beyond today for humanity to experience a better tomorrow. Space42 challenges traditional approaches with advanced AI and cutting-edge satellite technology, making space more accessible and redefining how data from space can be used on Earth. We aim to achieve this by connecting people to rewire potential, informing decisions to reimagine impact and enabling action to redefine tomorrow.

For more information visit: www.space42.ai, follow us on X and Instagram @Space42ai

Role Summary

As a Senior AI Engineer – Geospatial Intelligence & Advanced AI, you will lead the design, implementation, and operationalization of production-grade AI capabilities that turn multi-source geospatial data into decision-ready intelligence. You will work closely with Product Managers, Data Engineers, Software Developers, and domain experts to deliver scalable, reliable solutions across geospatial computer vision (e.g., satellite/remote-sensing imagery), graph/topology-aware modeling (networks and relational geospatial structures), optimization, and anomaly detection.

“Advanced AI” in this role is an accelerator: you will selectively apply modern AI patterns (including multimodal/foundation-model capabilities where appropriate) to improve analyst workflows, automation, and interpretability—while keeping GEOINT accuracy, traceability, and operational reliability as the primary success criteria.

You will be hands-on while also providing technical leadership—driving best practices for model development, ML system architecture, MLOps, reliability engineering, and responsible deployment.

Responsibilities

Technical leadership & delivery

• Lead end-to-end delivery of ML/AI solutions from problem framing and prototyping to production deployment, monitoring, and continuous improvement.
• Provide technical direction on model selection, experimentation strategy, evaluation methodology, and architecture.
• Mentor and review work of AI Engineers; raise engineering standards through code reviews, design reviews, and reusable patterns.
  
  

• Build AI-powered geospatial intelligence workflows that transform imagery and geospatial data into actionable outputs (alerts, change maps, object inventories, situational overlays).
• Partner with domain teams to define geospatial product requirements, acceptance criteria, and quality thresholds; translate them into model and pipeline requirements.
• Drive dataset strategy for GEOINT tasks (labeling guidelines, sampling, balancing, provenance, and ground-truth validation).
• Ensure model outputs are map-ready: georeferenced, interpretable, and compatible with cartographic and downstream GEOINT production pipelines (e.g., layers, metadata, confidence measures, and audit trails).
  
  

• Design, train, tune, and validate Deep Learning and classical ML models for geospatial computer vision (e.g., object detection/segmentation/change detection on satellite imagery).
• Develop and evaluate Graph Neural Network (GNN) models and topological learning approaches for problems involving spatial relationships, networks, and infrastructure graphs.
• Apply graph-based learning for use cases such as transport networks, connectivity analysis, supply chains, and geospatial relationship modeling.
• Contribute to analyst productivity and decision support by integrating advanced AI patterns where they measurably help (e.g., multimodal models for triage, structured extraction, search/summarization over GEOINT artifacts), with clear evaluation, governance, and guardrails.
• Define and implement evaluation approaches for advanced AI components (offline test sets, human-in-the-loop review, red-teaming where relevant) and ensure outputs are reliable, attributable, and policy-compliant.
• Build robust feature engineering pipelines and evaluation frameworks, including task-specific metrics (e.g., mAP/IoU, geospatial accuracy), graph metrics, and GenAI quality measures.
  
  

• Build and launch models in production using best practices in CI/CD, model versioning, experiment tracking, and automated testing.
• Define and manage SLAs/SLOs for model services (latency, throughput, accuracy drift, availability).
• Implement monitoring for data quality, model drift, and operational health; drive incident triage and resolution.
  
  

• Collaborate with engineers, product managers, and analysts to understand business needs and data requirements.
• Partner with data infrastructure teams to improve dataset reliability, lineage, governance, and access patterns.
• Build and maintain production-grade data pipelines (extraction, transformation, loading) that support training and inference.
  
  

• Develop clear documentation for models, pipelines, evaluation artifacts, and operational runbooks.
• Communicate trade-offs and results to technical and non-technical stakeholders.
• Enjoy collaborating in a multicultural and distributed environment.
  
  

• Master’s degree in Computer Science, Statistics, Information Systems, or another quantitative field (PhD is a plus).
  
  

• 5+ years of experience in software engineering, data engineering, or ML engineering (or equivalent).
• 3+ years of hands-on experience developing ML/AI algorithms and deploying them into production environments.
• Experience delivering GEOINT-grade analytics or products with quality controls (QA, traceability, auditability) and operational reliability expectations.
  
  

• Deep Learning frameworks (e.g., PyTorch, TensorFlow) and classical ML (e.g., scikit-learn, XGBoost).
• Geospatial intelligence & GeoAI: raster/vector data fundamentals, map projections/CRS, tiling/geo-referencing, and quality considerations for intelligence products.
• Geospatial computer vision: detection/segmentation/classification, change detection, multi-spectral/SAR basics, and post-processing for geospatial products.
• Familiarity with geospatial tooling and data stacks (e.g., GDAL/rasterio, GeoPandas equivalents, PostGIS, cloud-optimized formats such as COG, STAC concepts, and OGC-style interoperability is a plus).
• Graph AI and topology-aware learning: experience with Graph Neural Networks (GNNs), graph embeddings, network representation learning, and modeling relational/topological structures.
• Familiarity with graph ML frameworks such as PyTorch Geometric, DGL, or similar libraries.
• Familiarity with modern AI patterns for augmenting workflows (e.g., embeddings/vector search, basic multimodal model integration) and how to evaluate them safely in production.
• ML system design: online/offline inference, batch scoring, model serving patterns, feature stores.
• Data engineering: big data pipelines, architectures and datasets, queuing and streaming processing.
• Strong programming skills in Python and at least one of Java / Scala / Go.
• Version control and collaboration tooling (GitHub, GitLab).
  
  

• Experience with geospatial / remote sensing / satellite imagery data.
• Highly scalable systems and data stores.
• Cloud platforms such as AWS or GCP (experience with managed ML services is a plus).
• Cartography experience (map design principles, symbology, labeling, scale/CRS considerations) and delivering analyst-friendly geospatial products.
• Photogrammetry / 3D reconstruction experience (e.g., SfM/MVS concepts, orthorectification, DEM/DSM generation, bundle adjustment fundamentals) and working with aerial/satellite imagery pipelines.
• Experience translating AI outputs into cartographic deliverables (e.g., map layers, annotations, styling rules, and uncertainty visualization).
• Familiarity with geospatial tooling and formats (e.g., raster/vector concepts, GDAL/rasterio equivalents), and deployment of geospatial analytics at scale.
• Experience with MLOps tooling (e.g., MLflow, Kubeflow, Airflow, DVC, Weights & Biases).
• Containerization and orchestration (Docker, Kubernetes).
• LLM/VLM experience (nice-to-have): applying multimodal/LLM capabilities to GEOINT workflows (e.g., RAG over mission documentation, agentic tooling, structured extraction), plus practical evaluation and guardrails.
• Experience implementing GenAI safety patterns (PII handling, content filtering, jailbreak resistance), model governance, and security considerations.
• Experience building human-in-the-loop review workflows for GEOINT outputs (QA, adjudication, auditability) and operating under high-reliability expectations.