计算机视觉工程师

Job Description

At Terabase Energy, we believe that digitalization and automation will drive the next wave of innovation and cost reduction in large scale solar. To fully unlock the potential of this opportunity, Terabase is developing an interconnected software and construction automation platform.

We work alongside project developers, owners, and engineering & construction firms to support the design, optimization, and construction of huge solar projects around the world.

Our team is a blend of solar-industry veterans and novices, thought-leaders, dreamers, software, electrical and mechanical engineers, coders, product managers, project managers, sales and marketing professionals.

Terabase Energy is a well-funded scale-up headquartered in Berkeley, California, with several other offices around the world, with the most recent one being launched in Paris.

Our Paris office hosts our core R&D team in charge of the development of our flagship product, Construct.

Construct is a B2B SaaS construction management and quality control platform designed from the ground up for utility-scale solar projects. By combining GIS, AI/ML, IoT, and mobile technologies, the solution drives major improvements in safety, quality, cost, and schedule for the solar industry. Construct has supported more than 16+ GW of projects from leading owners and EPCs.

The Role

You'll join the Computer Vision Team, reporting to the Head of AI. You'll work at the intersection of deep learning, geospatial analysis, and real-world solar construction, building CV systems that are deployed at scale on active job sites around the world. This is a role for someone who moves fast, digs deep, and wants their work to have a direct impact on the energy transition.

The Computer Vision team owns the full ML lifecycle: from problem framing and dataset curation, through model development and validation, to production deployment and continuous improvement within Construct.

Beyond deep technical expertise, we're looking for someone who is genuinely excited about applying cutting-edge CV to one of the most impactful challenges of our time: accelerating the clean energy buildout.

You'll work closely with engineering teams in the Paris and our external development partners in Belgrade, Serbia.

What you'll do

• Write clean, production-grade Python code across the full ML stack, data pipelines, model training, inference services, and evaluation frameworks
• Track and apply the latest advances in computer vision and deep learning, from foundation models to efficient inference architectures, to real-world construction challenges.
• Design and run rigorous testing and validation protocols for ML-based products in challenging, real-world construction site conditions.
• Partner closely with cross-functional engineering teams (frontend, backend, data, infra) and external development partners to ship integrated features
• Own the end-to-end ML lifecycle: design, train, evaluate, and deploy models that digitalize every phase of a solar plant, from development and design through construction and operations.
• Serve as a technical bridge to users of developed tools, gathering field feedback, debugging edge cases, and closing the loop between model performance and real-world outcomes
• Contribute to team culture and cross-office cohesion, bridging the Paris R&D team with engineering, product, and field operations teams worldwide

Requirements

• Master's degree in Computer Science, Machine Learning, or a related field - or equivalent professional/research experience
• 2+ years of hands-on experience developing and deploying AI/ML models and pipelines (academic or industry)
• Strong Python proficiency, you write clean, testable, well-structured code
• Solid Computer Vision fundamentals and applied experience across detection, segmentation, classification, or related tasks
• Deep learning framework experience: PyTorch and/or TensorFlow
• Experience with data annotation tooling and labeling validation workflows
• MLOps experience: experiment tracking, model versioning, deployment pipelines (e.g. MLflow, DVC, Docker, CI/CD)
• Comfortable working remotely and collaborating asynchronously across time zones
• Strong written and verbal communication in English (French is a plus)
• Solid grasp of agile development practices and the full software product lifecycle

Nice to have

• PhD in Computer Science, Computer Vision, or a related field
• Experience with LLMs, Vision-Language Models, or Foundation Models applied to structured prediction or grounding tasks
• Track record building robust ETL pipelines for large, evolving datasets
• Experience with digital twin concepts, modeling and tracking physical assets and processes in software
• Hands-on experience with GIS technologies (PostGIS, GDAL, geospatial data formats, coordinate systems)
• Startup or scale-up experience, shipping new products in fast-moving, resource-constrained environments