Senior Software Engineer ML - Contractor position

About this opportunity

This is a future opportunity role, not a live start. We evaluate strong freelance ML engineers ahead of client demand so that when a project matches your profile, we come to you with concrete scope, rate and start date. Being accepted into the pool is not a guarantee of work, and we will be direct with you about timing: it may be weeks or it may be months. In exchange, we keep our process short, we do not ask for unpaid work, and once you are in, you hear about relevant projects before we go to market.

Engagement details

• Basis: Independent contractor. You invoice through your own company or registered sole-trader entity.
• Rate: Your quoted hourly or daily rate, in EUR or USD. Agreed in writing before any engagement starts.
• Commitment: Typically 40 hours per week unless specified differently, depending on the project.
• Duration: Typically 1–6 months per engagement. Extensions are common.
• Location: Fully remote. Most engagements require a minimum of four hours of daily overlap with US business hours.
• Working setup: You work as part of a Janea team on client projects, and are client-facing.

What engagements look like

• Taking an ML or LLM proof-of-concept to production, with the reliability and cost characteristics a large enterprise needs.
• Designing and hardening data and training pipelines for enterprise ML systems.
• ML platform and MLOps work: deployment, versioning, monitoring, evaluation, CI/CD for models.
• Building LLM and RAG systems, including retrieval quality, evaluation and cost control.
• Filling a specific, deep gap in a client engineering team for a defined period.
• Short technical audits and advisory work on an existing ML stack.

What we need from you

• 6+ years of software engineering experience, including 3+ years shipping ML systems into production.
• Strong Python, at production code quality. You are comfortable being the person responsible for what runs.
• Real depth in at least one of: LLM and agent systems; ML platform and MLOps; ML data and pipeline engineering — plus working competence in the other two.
• PyTorch or TensorFlow, and the production stack around them.
• Containers and Kubernetes in at least one major cloud (AWS, Azure or GCP).
• Workflow orchestration: Airflow, Kubeflow, Argo, Dagster or Prefect.
• Testing, benchmarking, CI/CD and monitoring practices applied to ML systems, not just to application code.
• Ability to turn an ambiguous business problem into a technical plan without close supervision.
• Client-facing English. You can explain a technical trade-off to a stakeholder who is not an ML specialist.
• Your own equipment and a legal entity to invoice from.

Useful, not required

• Agent frameworks: LangChain, LangGraph.