MLOps Engineer

Unico Connect is an AI-first product engineering agency that ships custom software, mobile apps, web platforms, and AI solutions for clients. We are hiring a Mid-Level MLOps Engineer to build and maintain the infrastructure, pipelines, and evaluation tooling behind the AI features in our client projects.

You will work alongside Senior MLOps Engineer and AI engineers, own execution on well-scoped problems, apply the security and testing patterns the senior team defines, and raise questions early when scope is unclear. This is hands-on: expect to write Terraform, wire CI, debug a broken inference endpoint, and ship a fix in the same day.

Responsibilities

• Training pipelines and data: Build reproducible pipelines for ingestion, preparation, and training, with data validation as code (Great Expectations, Soda, or Pandera).
• Experiment tracking: Configure MLflow or equivalent so teams can compare runs, identify best-performing models, and roll back when needed.
• Model deployment and serving: Package, deploy, and scale models on SageMaker, ECS, or EKS. Use vLLM, TGI, or BentoML where required.
• RAG and vector operations: Implement vector database integrations (Pinecone, Weaviate, Qdrant, pgvector, or OpenSearch), embedding pipelines, and retrieval workflows.
• Monitoring and feedback: Set up dashboards and alerts for model performance, drift, inference health, and cost anomalies. Support retraining triggered by drift or quality thresholds.
• Testing and evaluation: Run evaluation harnesses (Ragas, DeepEval, Promptfoo) in CI. Maintain golden datasets. Configure shadow and canary deployments.
• AI security basics: Apply guardrails (Llama Guard, NeMo Guardrails), PII redaction, prompt hardening, and secret management within the OWASP LLM Top 10 framework.
• CI/CD and infrastructure as code: Build CI/CD pipelines and Terraform modules across model, infrastructure, and application layers, with security and evaluation gates in the build.
• Troubleshooting and support: Debug production issues, optimise performance, and support client delivery teams. Participate in an on-call rotation as the team grows.

Requirements

• 3 years of hands-on MLOps experience (not adjacent roles like pure DevOps or data engineering).
• AWS: SageMaker, ECS or EKS, S3, Lambda, Step Functions, CloudWatch. Azure or GCP experience acceptable with willingness to work in AWS.
• Docker, Kubernetes fundamentals, Terraform, and at least one CI/CD platform (GitHub Actions, Jenkins, or GitLab CI).
• MLflow or equivalent, one pipeline orchestrator (Airflow, SageMaker Pipelines, or Kubeflow), and exposure to monitoring tools (Evidently, WhyLabs, Arize, or Langfuse).
• GenAI hands-on: at least one vector database, one LLM framework (LangChain or LlamaIndex), and practical RAG experience in production.
• Data validation in production pipelines (Great Expectations, Soda, or Pandera).
• Working knowledge of LLM security: prompt injection, PII handling, secrets management, and OWASP LLM Top 10.
• Solid Python, Bash scripting, basic SQL, and object storage (S3 or equivalent).