MLOps Engineer (AI / LLM Platforms, MedTech)

We help clients design and build AI-driven products and platforms. You’ll work across projects implementing RAG, real-time chat, and LLM integration for enterprise SaaS applications. This involves multi-tenant architectures, regulatory and compliance workflows, and integrating AI services with existing backend systems.

Responsibilities

• Design and implement AI service layers for enterprise SaaS products: RAG, embeddings, and LLM orchestration

• Build real-time chat interfaces with context-aware retrieval (documents, structured data, vector search)

• Implement data pipelines for ingestion, normalization, and AI enrichment

• Design multi-tenant data isolation, cost tracking, logging, and auditability

• Integrate LLM providers (e.g., AWS Bedrock, OpenAI) with vector stores, relational databases, and object storage

• Work with Kubernetes, message queues, and async processing patterns

• Collaborate with client and internal teams on requirements, security, and governance

Required experience

• 5+ years of backend development, including production systems using AI/LLM

• Strong PHP and Laravel: APIs, queues, jobs, and background processing

• Experience with RAG: vector search, embeddings, and retrieval design

• Experience integrating LLM APIs and designing prompts

• Experience with cloud providers (e.g., AWS) and containerization (Docker, Kubernetes)

• Experience with relational databases (PostgreSQL/MySQL), object storage, and message queues

• Experience designing multi-tenant systems with clear data separation

• Familiarity with structured logging, metrics, and monitoring

Preferred experience

• Vector databases (OpenSearch, Pinecone, or similar)

• Document processing and OCR pipelines

• Workflow or job orchestration tools

• Regulated or compliance-sensitive domains (healthcare, finance)

Tech stack (typical)

• Backend: PHP, Laravel

• Cloud: AWS or equivalent (managed Kubernetes, RDS, object storage)

• Data: PostgreSQL/MySQL, vector DB, Redis

• Infrastructure: Kubernetes, Docker, Terraform or similar

• Observability: Prometheus, Grafana, structured logging