Are you building AI agentsnot just API wrappers? Were looking for an Agentic AI Engineer who loves designing intelligent, production-grade systems and turning LLM capabilities into reliable enterprise solutions. What You’ll Work On Architect and build single & multi-agent systems with complex orchestration and workflow design.

Develop production-grade RAG pipelines, semantic search, vector embeddings, and vector databases. Build intelligent workflows covering data ingestion retrieval reasoning tool calling inference response. Engineer and optimize prompts, context strategies, and agent workflows for accuracy, reliability, and performance.

Design custom domain-specific guardrails to handle complex and non-trivial business scenarios. Build robust LLM evaluations (Evals) to measure quality, accuracy, hallucination, and task performance. Debug and trace agentic workflows using MLflow, LangSmith, or similar observability/evaluation platforms.

Continuously evaluate emerging AI frameworks and technologies and bring the right ones into production.





Tech Stack We’re Looking For

Years of Experience - 3 to 6Years

Location - Hyderabad (WF0)

Mode of Interview - 3 rounds - Last round is F2F from Hyd

Must Have

- Python | LangChain | LangGraph | LLMs & Prompt Engineering RAG | Vector Embeddings & Semantic Search | Agent Engineering Multi-Agent Orchestration | Custom Guardrails | LLM Evals LangSmith / MLflow / Similar AI Observability Tools You'll Be a Excellent Fit If You Have 3Years+ of experience across Data Science, ML, AI Engineering, or related domains.
- Have actually built and shipped agentic systems, not just experimented with LLM APIs.
- Can think beyond individual models and design end-to-end AI system architectures.
- Understand the trade-offs between accuracy, latency, cost, reliability, and scalability.
- Enjoy solving ambiguous problems and turning them into production-ready AI solutions.

📌 Agentic AI Engineer (Hyderabad)
🏢 CoffeeBeans Consulting
📍 Hyderabad