Lead Agentic AI Engineer

Syren Cloud Inc.; Hyderabad; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

Syren Cloud Inc. designs and builds autonomous AI agents and multi-agent systems, turning language models into practical agents that plan, use tools, and complete real tasks end to end.

You will architect agent frameworks, design orchestration layers, and optimize RAG pipelines, while collaborating with product teams to meet latency, cost, and reliability targets.

Qualifications

• 3+ years building software, with 1+ year hands‑on building LLM-powered agents or applications.
• Strong Python; solid grasp of prompt engineering and LLM behavior.
• Hands‑on experience with at least one agent framework (LangGraph, CrewAI, AutoGen, or equivalent).
• Practical experience with tool‑use/function‑calling and RAG pipelines and vector databases.
• Ability to debug non‑deterministic systems and design for failure and reliability.
• Open‑source contributions to agent or LLM tooling.
• Experience with multi‑agent orchestration at scale (concurrency, cost control, observability).
• Fine‑tuning or model‑adaption experience.

Responsibilities

• Architect and build single- and multi-agent systems using frameworks like LangChain/LangGraph, CrewAI, and AutoGen.
• Design orchestration layers: task planning, agent routing, memory, state management, and error/recovery handling.
• Implement robust tool-use and function-calling — connecting agents to APIs, databases, and internal services.
• Build and optimize RAG pipelines (retrieval, chunking, re-ranking, grounding) to reduce hallucination and improve task accuracy.
• Own agent evaluation and iteration: define success metrics, run experiments, and harden agents against edge cases.
• Collaborate with product and platform teams to ship agents that meet latency, cost, and reliability targets.

Skills

Python

Prompt engineering

LLM-powered agents

Agent frameworks

Debugging non-deterministic systems

Open-source contributions

Multi-agent orchestration

Model adaptation

Tools

LangGraph

CrewAI

AutoGen

LangChain

Vector databases

Job description

Summary

Design and build autonomous AI agents, multi-agent systems, and the orchestration layers that make them reliable in production. You\'ll turn LLM capabilities into agents that plan, use tools, and complete real tasks end to end.

• Architect and build single- and multi-agent systems using frameworks like LangChain/LangGraph, CrewAI, and AutoGen.
• Design orchestration layers: task planning, agent routing, memory, state management, and error/recovery handling.
• Implement robust tool-use and function-calling — connecting agents to APIs, databases, and internal services.
• Build and optimize RAG pipelines (retrieval, chunking, re-ranking, grounding) to reduce hallucination and improve task accuracy.
• Own agent evaluation and iteration: define success metrics, run experiments, and harden agents against edge cases.
• Collaborate with product and platform teams to ship agents that meet latency, cost, and reliability targets.

Required Skills

• 3+ years building software, with 1+ year hands‑on building LLM-powered agents or applications.
• Strong Python; solid grasp of prompt engineering and LLM behavior.
• Hands‑on experience with at least one agent framework (LangGraph, CrewAI, AutoGen, or equivalent).
• Practical experience with tool‑use/function‑calling and RAG pipelines and vector databases.
• Ability to debug non‑deterministic systems and design for failure and reliability.
• Open‑source contributions to agent or LLM tooling.
• Experience with multi‑agent orchestration at scale (concurrency, cost control, observability).
• Fine‑tuning or model‑adaption experience.

Core Stack: