Senior AI Platform Engineer - LLM/RAG

 Design, build, and maintain AI agent workflows and orchestration patterns using frameworks like LangGraph.

 

- Build and optimize production-grade RAG systems chunking strategies, retrieval pipelines, embedding generation, and response quality.

 

- Implement and manage LLM API integrations with proper retry logic, fallbacks, rate limiting, error handling, and cost optimization.

 

- Develop prompt engineering solutions including system prompts, few-shot patterns, structured outputs (JSON mode), and prompt versioning.

 

- Build input validation, output filtering, and content safety layers for production AI systems.

 

- Implement AI observability and evaluation tracing, automated quality checks, regression testing for AI outputs.

 

- Build data ingestion pipelines for AI systems (document processing, embedding generation, vector storage).

 

- Collaborate with architects and product teams to translate AI capabilities into reliable platform features.

 

- Write well-tested, production-ready code with unit tests, integration tests, and AI-specific tests.

 

- Contribute to design docs, runbooks, and technical documentation for AI systems.

 

Requirements :

 

Must Have :

 

- 5+ years of professional software development experience.

 

- Hands-on experience building AI/ML features or LLM-powered systems in production.

 

- Strong understanding of LLM fundamentals tokens, context windows, embeddings, temperature, and similarity search.

 

- Experience with RAG systems indexing strategies, retrieval methods, chunking, and when RAG is the right approach.

 

- Practical prompt engineering skills chain-of-thought, few-shot, structured outputs, and systematic iteration.

 

- Experience with at least one AI orchestration framework (LangGraph, LangChain, or equivalent).

 

- Working experience with at least one vector database (pgvector, Pinecone, Qdrant, Weaviate, or similar).

 

- Proficiency in Python for AI development and prototyping.

 

- Experience with LLM APIs (Claude SDK, OpenAI SDK, or similar) in production.

 

- Understanding of AI safety basics prompt injection, jailbreaking, and practical mitigation approaches.

 

- Working knowledge of cloud platforms (AWS preferred) and containerization (Docker, Kubernetes).

 

- Strong problem-solving skills and ability to break down ambiguous AI problems into implementable tasks.

 

Nice to Have :

 

- Experience with AI observability tools (LangSmith, LangFuse) for tracing and debugging.

 

- Experience with MCP (Model Context Protocol) or equivalent tool integration patterns.

 

- Familiarity with fine-tuning concepts and when to consider fine-tuning vs. prompt-based solutions.

 

- Programming experience in Java

 

- Experience with event-driven architectures and message queues (Kafka, RabbitMQ).

 

- Knowledge of the telecom domain (billing, CRM, customer lifecycle).

 

- Experience shipping AI features in a customer-facing product.

 

- Backend experience in Go, Node.js, or Java microservices.