Job Opening For Agentic AI Lead - Internal Project

Position: Agentic AI Lead
Location: Bangalore/Salem
No of Position: 2
Experience: 5-8 Yrs.

Core Competencies:

• Large language model (LLM) mastery: Deep understanding of how LLMs work, their limitations (e.g., hallucinations, biases), and how to apply models from providers like OpenAI, Anthropic, Google, and open-source models like LLaMA.
• Agentic frameworks and architecture: Proficiency with specialized tools for building and orchestrating multi-agent systems, such as:
  • LangChain/LangGraph: Used for creating chains of tasks and multi-agent workflows.
  • AutoGen: A framework by Microsoft that simplifies multi-agent communication and collaboration.
  • CrewAI: Ideal for assigning specific roles and personas to different AI agents working together.
• Retrieval-Augmented Generation (RAG): Expertise in implementing RAG systems, which allow agents to retrieve information from large, domain-specific knowledge bases to generate more accurate and relevant responses.
• Tool integration and function calling: The ability to teach AI agents how to interact with external software, APIs, and databases (e.g., sending emails, querying a CRM) to expand their capabilities.
• Deep learning and machine learning: A strong foundation in ML fundamentals, including supervised, unsupervised, and reinforcement learning, with a specific focus on deep learning techniques relevant to generative models.

Tools & Technologies: GraphRAG, Claude, Llama, CrewAI, LangGraph, LangChainOpenSearch, AWS SageMaker, AWS Bedrock, Neo4j, Jupyter Notebooks
Programming Languages: Python, R, SQL
Deployment & Integration: Model Deployment, RESTful APIs, Docker, Kubernetes, CI/CD Pipelines, Cloud Services Integration
Methodologies: Generative AI, Agentic AI, Transfer Learning, Fine Tuning
Optimization & Performance: Model Optimization, A/B Testing, Scalability, Performance Tuning, Distributed Computing
Soft Skills: Problem Solving, Critical Thinking, Collaboration, Communication, Continuous Learning

Job Responsibilities:

• Agentic AI Leader will be responsible for research & innovation, technology development including quick architecture, design & development of prototypes and ensuring timely evolution of prototypes into products. The primary objective of this role will be to ensure the constant launch of innovative products for consumers with the help of a small team of highly qualified research technologists / developers.
• Build and lead a team of Agentic AI leads, data scientists, Agentic AI engineers, and researchers, fostering a collaborative environment and technical growth.
• Oversee the full project lifecycle for Agentic AI applications, from ideation and design to implementation and evaluation, ensuring timely delivery of high-quality results..
• Evaluate the performance of AI models and make necessary improvements.
• Design autonomous and multi-agent systems: Architect complex systems where multiple AI agents can collaborate, delegate tasks, and communicate with one another to achieve a shared goal.
• Develop agent frameworks: Build systems with core agentic capabilities, such as perception, reasoning, planning, memory management, and self-correction.
• Integrate LLMs: Use large language models (LLMs) from providers like OpenAI, Google, and Anthropic as the "brain" for the agents, customizing their behavior with specific prompts.
• Implement Retrieval-Augmented Generation (RAG): Create data pipelines that give agents access to vast, domain-specific knowledge bases, such as internal documents and enterprise data, for more accurate and grounded responses.
• Connect agents to tools and APIs: Enable agents to interact with the real world by giving them the ability to call external functions, use web search, query databases, and connect to enterprise systems like CRMs or ERPs.
• Build full-stack agentic applications: Develop complete applications, from the back-end services that power the agents to the front-end user interfaces that allow for dynamic, conversational interaction.
• Lead the Deployment of AI workflows: Use cloud platforms such as AWS, GCP, and Azure to deploy AI systems at scale, optimizing for GPU efficiency, cost, and latency.
• Implement observability and monitoring: Set up tools (e.g., LangSmith) to track and refine agent behavior, ensuring predictable and reliable performance in production.
• Manage model lifecycle: Oversee the entire process from training to deployment, including regular updates and retraining of models as needed.
• Implement measures to monitor and address risks related to bias and performance in generative AI models