Design and build agentic systems: single agents, multi-agent orchestration, and sub-agent patterns for complex workflows (planner/executor, supervisor/worker, hierarchical task decomposition)
Develop MCP (Model Context Protocol) servers and tools that expose internal systems, datasets, and actions to LLM-powered applications in a safe, governed, and reusable way
Implement retrieval systems (RAG, hybrid search, graph-based retrieval) including chunking, embedding, re-ranking, and context-assembly strategies
Build and maintain MLOps automation: CI/CD for models and agents, environment management, artifact handling, and versioning of prompts, models, data, and code
Implement observability for AI systems: tracing, token/latency/cost metrics, quality and drift monitoring, alerting, and incident response
Build and maintain data pipelines for ingestion, transformation, and export across multiple sources and destinations
Expose well-modelled, governed datasets and APIs that agents, tools, and downstream consumers can rely on
Ensure secure data handling and compliance with relevant data protection standards and internal policies
Contribute to documentation, standards, and continuous improvement of the data platform and engineering processes

Bachelor’s degree in Computer Science, Engineering, Mathematics, or a related technical field (or equivalent practical experience)
5+ years of Data or ML Engineering experience, with at least 3 years shipping AI or ML systems to production.
Strong Python skills (typed code, async, testing) and solid SQL fluency.
Hands-on experience building agentic applications with frameworks such as LangGraph, LlamaIndex, CrewAI, or the Anthropic/OpenAI Agents SDKs — including tool use, memory, and multi-step reasoning patterns.
Practical experience with MCP or comparable tool/function-calling protocols; comfortable designing tool schemas and sub-agent boundaries.
Experience with RAG architectures, vector stores (e.g. pgvector, Pinecone, Weaviate), and embedding models
Familiarity with at least one major cloud provider (GCP, AWS, Azure) and deploying data solutions in the cloud
Strong DevOps fundamentals: CI/CD (GitHub Actions, Cloud Build, or similar), IaC (Terraform), containerisation (Docker), and orchestration (Kubernetes or serverless equivalents)
Comfortable building and maintaining data pipelines with orchestrators (Airflow/Composer, Dagster) and distributed engines (Spark, BigQuery)
Strong troubleshooting mindset: ability to debug issues across data, infra, pipelines, and deployments.
Collaborative mindset and clear communication across engineering, analytics, and business stakeholders.

AI Engineer – Contractor

Key skills