AI Archive API Developer (FastAPI/AWS/Computer vision) - July 2026

<span><span><span>For one of our clients in the sportswear industry we are looking for a freelance <strong>AI Archive API Developer (FastAPI/AWS/Computer vision)</strong></span></span></span><br /> <br /> <br /> <b><span lang="en" xml:lang="en"><span><span>Overview:</span></span></span></b> <ul> <li><span lang="en" xml:lang="en"><span><span>Short description of the project: The AI Archive project is an initiative within the client to research, prototype, and develop internal capabilities for 3D product design and development using generative AI. The project focuses on identifying, adapting, and fine-tuning state-of-the-art 3D generative models to support client-specific use cases.</span></span></span></li> <li><span lang="en" xml:lang="en"><span><span>The toolchain leverages modern machine learning architectures and in-house datasets to enable AI assisted product design and engineering/creation. </span></span></span></li> </ul> <br /> <b><span lang="en" xml:lang="en"><span><span>Tasks:</span></span></span></b><br /> <b><span lang="en" xml:lang="en"><span><span>Implementation of data preparation pipelines for model training: </span></span></span></b> <ul> <li><span lang="en" xml:lang="en"><span><span>Processing and preparation of multimodal data: images, segmentation masks, depth data, and 3D assets (meshes, materials, textures) tailored to the specific technical requirements of each generative AI use case.</span></span></span></li> <li><span lang="en" xml:lang="en"><span><span>Preparation and validation of training datasets across modalities. Validation against technical requirements and shared upfront data quality.</span></span></span></li> </ul> <b><span lang="en" xml:lang="en"><span><span>Implementation of backend services and APIs: </span></span></span></b> <ul> <li><span lang="en" xml:lang="en"><span><span>Development of REST APIs using Ray Serve for model inference endpoints.</span></span></span></li> <li><span lang="en" xml:lang="en"><span><span>Integration with SQL databases for data access, persistence, and API backing.</span></span></span></li> <li><span lang="en" xml:lang="en"><span><span>Deployment as standalone jobs, backend service integrations, or cloud-based execution on AWS.</span></span></span></li> <li><span lang="en" xml:lang="en"><span><span>Utilization of AWS services (S3, EC2, IAM) for storage, compute and permissions.</span></span></span></li> <li><span lang="en" xml:lang="en"><span><span>Utilization of AWS Bedrock for managed model inference and AWS agentcore for building and deploying AI agent workloads.</span></span></span></li> </ul> <br /> <b><span lang="en" xml:lang="en"><span><span>Execution of machine learning workflows and environment configuration:</span></span></span></b> <ul> <li><span lang="en" xml:lang="en"><span><span>Running inference workloads using the PyTorch framework.</span></span></span></li> <li><span lang="en" xml:lang="en"><span><span>Independent configuration and management of Python environments and dependencies across all deployment targets.</span></span></span></li> </ul> <p><br /> <b><span><span><span>Start:</span></span></span></b><span><span><span> ASAP<br /> <b>Capacity: </b>full-time, 40h/week<br /> <b>Duration: </b>till end of 2026<br /> <b>Location: </b>preference for 50% onsite (Herzogenaurach), if not 100% remote is also OK</span></span></span><br />  </p>