Responsibilities
- Build and maintain scalable backend services using modern technologies such as Python, Node.js, Java, or Go.
- Design and implement APIs that expose AI/LLM-powered capabilities to web, mobile, and enterprise applications.
- Integrate production-grade LLM APIs (OpenAI, Anthropic, Gemini, Azure OpenAI, etc.) into backend systems and workflows.
- Build and manage Retrieval-Augmented Generation (RAG) pipelines including ingestion, chunking, embedding, indexing, retrieval, reranking, and grounding.
- Design and maintain enterprise knowledge bases optimized for LLM and agent consumption.
- Build agentic workflows and multi-step reasoning systems using frameworks such as LangGraph, CrewAI, AutoGen, or equivalent.
- Design AI-enabled automation flows using tools such as n8n, Temporal, Airflow, or similar orchestration platforms.
- Build internal AI-powered developer productivity tools including code assistants, automated documentation generators, test generation systems, release-note generators, and incident-analysis agents.
- Handle LLM operational concerns including prompt management, context engineering, latency optimization, retries, fallback strategies, caching, observability, and cost optimization.
- Work with vector databases and search platforms such as Pinecone, Weaviate, Qdrant, Elasticsearch, or FAISS.
- Implement secure tool integrations and MCP (Model Context Protocol)-based workflows connecting APIs, databases, and enterprise systems to AI agents.
- Design monitoring and evaluation pipelines for AI systems including tracing, prompt/version tracking, hallucination analysis, token/cost monitoring, and performance evaluation.
- Collaborate closely with frontend, mobile, DevOps, QA, security, and product teams in a structured engineering environment.
- Contribute reusable SDKs, internal frameworks, and shared AI platform components.
Skills Required
Backend Engineering
- Strong backend development experience in Python, Node.js, Java, or Go
- REST / GraphQL API design and development
- SQL and NoSQL databases
- Distributed systems and asynchronous processing
- Queues and event-driven architectures (Kafka, RabbitMQ, Pub/Sub, etc.)
AI / LLM Engineering
- Production experience integrating OpenAI, Anthropic, Gemini, or equivalent LLM APIs
- Prompt engineering and context management
- RAG architecture and retrieval pipelines
- Knowledge base construction for LLM systems
- Embedding strategies:
- Dense embeddings
- Sparse retrieval
- Hybrid search
- Reranking pipelines
- Multilingual/domain-specific embeddings
- Vector databases and semantic search systems
Agentic AI & Workflow Orchestration
- LangGraph / CrewAI / AutoGen or similar agent frameworks
- n8n / Temporal / Airflow or equivalent orchestration systems
- MCP (Model Context Protocol) awareness and tool integration patterns
- Agent memory, tool-calling, and workflow design fundamentals
Observability & Reliability
- AI observability and tracing tools such as LangSmith, Langfuse, MLflow, OpenTelemetry, Grafana, Datadog, or equivalent
- Token, latency, retry, and cost monitoring
- Evaluation pipelines for prompts and agent workflows
- Logging, metrics, tracing, and production debugging
Cloud & DevOps
- AWS / GCP / Azure
- Docker and containerized deployments
- CI/CD pipelines
- Automated testing and release workflows
Ideal Profile
- 4–7 years of backend engineering experience with strong system design fundamentals.
- Hands-on experience building AI-powered backend systems, RAG pipelines, or agentic workflows in production environments.
- Strong understanding of LLM limitations, hallucination mitigation, grounding strategies, and cost-performance tradeoffs.
- Experience designing scalable AI infrastructure and enterprise-grade APIs.
- Familiarity with agent monitoring, AI evaluation frameworks, and workflow orchestration platforms.
- Strong debugging, performance optimization, and problem-solving skills.
- Experience working in cross-functional product and engineering teams.
