Software Engineer

Capgemini; Bengaluru Urban; On-site; INR 1,500,000 - 2,500,000; Full time

Job summary

A leading IT consulting firm in Bengaluru is seeking a Mid-Senior level professional to design and develop Agentic AI architectures. The candidate will work with Large Language Models and must have experience in RAG architectures and model fine-tuning. This is a full-time position that offers an exciting opportunity in the rapidly evolving field of AI.

Qualifications

• Experience in developing and deploying Agentic AI systems.
• Hands-on experience with RAG architectures and related tools.
• Knowledge in model fine-tuning using frameworks like HuggingFace Transformers.

Responsibilities

• Design and develop Agentic AI architectures that can autonomously plan, reason, and execute tasks.
• Implement multi-agent communication protocols for agent-to-agent collaboration.
• Develop and integrate RAG pipelines to enhance reasoning capabilities.

Skills

Developing Agentic AI systems

Experience with LLMs

RAG architectures

Python proficiency

Tools

HuggingFace Transformers

PyTorch

TensorFlow

Job description

Responsibilities

Design and develop Agentic AI architectures that can autonomously plan, reason, and execute tasks.

Implement multi-agent communication protocols for agent-to-agent collaboration and coordination.

Work with Large Language Models (LLMs) such as LLaMA, GPT, etc., for language understanding, generation, and task planning.

Develop and integrate Retrieval-Augmented Generation (RAG) pipelines to enhance the reasoning capability of agents with external knowledge.

Perform fine-tuning of foundational models for specific domains, tasks, or use cases.

Design and experiment with lightweight models (e.g., Phi, Tiny LLMs) for efficiency in real-time or on-device scenarios.

Collaborate cross-functionally with Data Scientists, ML Engineers, and Product Teams to deliver end-to-end AI solutions.

Conduct rigorous testing and evaluation of agent behaviors, decision-making, and performance.

Must-Have Qualifications

• Proven experience in developing and deploying Agentic AI systems.
• Strong experience with LLMs, particularly Meta’s LLaMA family or similar open-source models.
• Hands-on experience with RAG (Retrieval-Augmented Generation) architectures and related tools like LangChain, Haystack, etc.
• Knowledge and experience with model fine-tuning using frameworks like HuggingFace Transformers, PEFT, or LoRA.
• Experience in implementing agent-to-agent communication frameworks and multi-agent task handling systems.
• Proficiency in Python and AI/ML libraries such as PyTorch, TensorFlow, Transformers, etc.
• Familiarity with prompt engineering and chaining logic for LLMs.

Seniority level

• Mid-Senior level

Employment type

• Full-time

Job function

• Information Technology

Industries

• IT Services and IT Consulting