AI Backend Engineer - Hybrid CDMX

Our client is looking for a Principal Backend Engineer to design and build the backend and GraphQL API architecture for AI-powered retrieval systems (RAG) and AI agent orchestration. You´ll develop high-performance GraphQL APIs that serve LLM-driven search, vector retrieval, and graph-based recommendations to users in real time. You´ll also develop frameworks and orchestration to coordinate autonomous AI agents at scale.

This role requires expertise in those areas to help architect and implement the right solutions for AI-driven retrieval. You´ll work closely with the Head of Engineering to design scalable AI-powered APIs and retrieval pipelines while remaining hands-on with coding.

Tech Stack: Golang (preferred), Python, GraphQL, PostgreSQL, Redis, Kubernetes, Azure, Vector Databases (Weaviate, Pinecone, FAISS), Graph Databases (Neo4j, AWS Neptune, TigerGraph), AI Agent frameworks (LangGraph, AutoGen, CrewAI, etc)

Why This Role?

• Architect + Code You'll actively design GraphQL APIs for AI retrieval and an AI agent platform while writing production-grade code daily.
• AI-Powered Search & Retrieval (RAG) Leverage your experience with vector search and knowledge graphs to define future system architecture.
• Scalability & Performance Ensure low-latency, high-scale orchestration and retrieval in a GraphQL-first environment.

Responsibilities

• Design, build, and optimize GraphQL APIs to serve AI-powered search and retrieval systems.
• Design, build and optimize AI agent platform and framework to orchestrate autonomous agents.
• Hands-on coding (70%), focusing on Golang & Python API development.
• Architect GraphQL queries, mutations, and resolvers to support LLM-powered recommendations.
• Optimize GraphQL query performance (batching, pagination, caching, Dataloader optimizations).
• Guide and implement vector search and knowledge graph capabilities (Weaviate, Pinecone, FAISS, Neo4j, AWS Neptune, TigerGraph).
• Expose RAG and GraphRAG retrieval systems through GraphQL API endpoints.
• Database design for customer-facing access patterns.
• Ensure GraphQL schema design is flexible, scalable, and AI-friendly.
• Implement security best practices (OAuth, JWT, role-based access in GraphQL).
• Collaborate with ML/AI engineers to expose LLM models, embeddings, and knowledge graphs via APIs.
• Deploy and scale services in Azure (Kubernetes, Terraform, AKS).

What Were Looking For

• 5+ years of backend development experience, with strong Golang (preferred) and Python skills.
• Proven experience designing and optimizing GraphQL APIs for high-performance applications.
• Strong understanding of AI retrieval systems, including RAG, GraphRAG, vector search, and knowledge graphs (even if not currently in use).
• Exposure to AI agent frameworks and design patterns.
• Deep expertise in distributed systems, microservices, and GraphQL performance tuning.
• Experience integrating AI-driven APIs with GraphQL queries and resolvers.
• Azure cloud experience (AKS, Functions, Blob Storage, CosmosDB).
• Proven track record of hands-on coding while also defining backend architecture and best practices.

Bonus Points

• Experience with GraphQL federation, schema stitching, or Apollo Gateway.
• Familiarity with GraphQL + WebSockets (subscriptions for real-time AI updates).
• Exposure to MLOps and model-serving platforms (AWS SageMaker/Bedrock, ClearML, Triton).