Senior Backend Engineer (with ML experience)

Remote | 7000 - 10000$ (Gross)

Company: AGIMA Location: not RU/BY (Russia/Belarus) Format: remote Employment: full-time, project-based (until the end of the year with possible extension) Salary range: 7-10 000$ (Gross)

We are looking for a backend engineer who will be responsible for creating and operating systems that enable deploying Generative Artificial Intelligence (GenAI) into an industrial environment for processing documents and blueprints (text and images). Your role involves designing scalable microservices, APIs, and inference pipelines that interface with foundational models (currently we use AWS Bedrock) and making their functionality reliable and ready for industrial operation.

Responsibilities: • Design and implement backend microservices and APIs that integrate with foundational models for processing documents and blueprints. • Build scalable request routing strategies, caching, and inference queues (capable of handling sudden load spikes). • Implement CI/CD and monitoring mechanisms for inference pipelines (logging, metrics, alerts). • Deploy and operate services in Kubernetes (responsible for deployment, rollout, and health checks; cluster administration is provided by the platform/SRE team). • Integrate services with AWS services and ensure proper access identity (IAM) and least privilege for inter-service communication. • Collaborate with data scientists and machine learning engineers (they are responsible for model training; you are responsible for production integration). • Establish operational practices for inference in production (the scope of on-call tasks will be specified; expected incident response at the service level).

Requirements: • Minimum 5 years of experience building industrial backend systems and APIs in Python. • Experience designing and operating distributed microservices (scalability, routing, caching, failure modes). • Practical experience with AWS and experience integrating external model APIs (currently we primarily interface with foundational models via Bedrock). • Experience with industrial operation of services in Kubernetes (you will be responsible for application deployment and operation; cluster infrastructure is maintained by the platform team). • Experience with CI/CD (knowledge of GitHub Actions is required). • Solid practical knowledge of IAM and core cloud security principles. • Clear understanding of logging and observability and the ability to design pipelines for subsequent analysis. • Experience with Terraform or other Infrastructure as Code (IaC) tools. • Ability to clearly and convincingly communicate system design decisions and previous projects.

Nice to have: • Previous experience integrating GenAI and LLMs (Bedrock, API routing, model selection). • Experience with image and document processing pipelines (OCR, image preprocessing, technical drawing recognition). • Experience with inference orchestration patterns (async workers, batching, GPU orchestration) or familiarity with GPU workloads. • Familiarity with ML lifecycle tools (MLflow, model registries, experiment tracking). • Experience with real-time and batch inference services, feature stores, or experience using a SageMaker production environment.

*The role involves working in an English-speaking team, in Germany's time zone (CET/CEST).

To apply: @staff_eeva