We are looking for LLM Engineers to design and build the reasoning, tool-use, and chaining logic that powers our AI framework. This role focuses on how large language models think, plan, decide, and act—including prompt strategies, reasoning chains, tool invocation, and structured outputs.

Job Title: LLM Engineer – Agent Reasoning & Tooling

Location: Bangalore, Hyderabad, Pune

Experience: 5–10+ years overall experience, with strong hands-on depth in LLM-based systems

Key Responsibilities:

• Design and implement reasoning chains (plan–execute–reflect, ReAct, self-consistency, decomposition strategies)
• Build structured chain-of-thought abstractions without exposing internal reasoning unnecessarily
• Optimize reasoning flows for accuracy, latency, and cost
• Translate architectural patterns into concrete agent behaviors
• Implement memory usage patterns (short-term, long-term, contextual retrieval)
• Support multi-agent interactions through well-defined interfaces and protocols

Required Skills

• Hands-on experience with LLMs and agent frameworks such as GPT, Claude, Mistral, or similar
• Strong programming skills in Python (or equivalent)
• Strong understanding of reasoning patterns (ReAct, CoT, Tree-of-Thoughts, planner–executor)
• Experience implementing tool-augmented LLM workflows
• Experience with APIs, async workflows, and distributed systems concepts
• Ability to write clean, testable, and maintainable code
• Experience with LangChain, Semantic Kernel, AutoGen, CrewAI, or similar
• Experience with RAG, vector databases, or hybrid retrieval approaches
• Strong Python programming with experience in Hugging Face transformers, LangChain,OpenAI SDK