Must Have Skills :

- 2+ years of hands-on experience in agentic AI / multi-agent systems.

- Proficiency with LangChain, Langraph, CrewAI, AutoGen, Haystack, or equivalent frameworks.

- Strong background in Python and experience with prompt engineering, tools integration, and chaining logic.

- Solid understanding of LLM APIs, RAG, vector stores, tool use, and memory architectures.

- Hands-on experience with open-source and commercial LLMs (e.g., GPT-4, Claude, Gemini, Mistral).

- Experience deploying AI agents in cloud-native environments (AWS, GCP, Azure).

- Ability to lead architectural discussions, PoCs, and hands-on development in fast-paced environments.

- Model-cost profiling and budgeting (API call minimization, batch vs. streaming)

- Latency tuning for real-time agents, Autoscaling strategies.

- Exposure to Autonomous AI agents (AutoGPT, BabyAGI, CAMEL, MetaGPT).

- Understanding of LLM fine-tuning, adapters, and RLHF.

- Experience with agent simulation, environment modeling, or reinforcement learning is a plus.

- Familiarity with compliance, privacy, and safety in GenAI deployments.

- Prior experience in building domain-specific agents (Lifescience, healthcare, Pharma).

Key Responsibilities :

- Architect and implement agentic AI systems using modern LLM orchestration frameworks (LangChain, CrewAI, AutoGen, etc.)

- Design multi-agent collaboration models including planner-solver, autonomous teams, and goal decomposition agents.

- Build reusable tooling, APIs, and memory architectures for agent interaction, coordination, and context persistence.

- Lead hands-on development and deployment of GenAI applications (e.g., assistants, copilots, decision support).

- Evaluate and integrate LLMs (OpenAI, Claude, Mistral, LLaMA, etc.), vector databases (Pinecone, Weaviate, FAISS), and retrieval systems (RAG).

- Optimize agent performance for real-time environments, reliability, scalability, and ethical constraints.

- Guide teams in adopting agent frameworks, best practices, prompt engineering, and model fine-tuning.

- Collaborate with stakeholders to translate business requirements into technical solutions using agent-based paradigms.

- Continuously monitor trends in multi-agent systems, cognitive architectures, and open-source AI framewor