Manager | AI Engineer | Bengaluru | Control Assurance

Summary: Design, develop, and deploy Generative AI solutions leveraging Large Language Models and RAG pipelines. Manage embedding-based search systems, optimize prompts, and integrate LLMs via APIs for production-grade applications within the Control Assurance team.

Manager | AI Engineer | Bengaluru | Control Assurance
• Job requisition ID : 110969
• Location: Bengaluru
• Entity: Deloitte Touche Tohmatsu India LLP

The Team

Assurance is about much more than just the numbers. It’s about attesting to accomplishments and challenges and helping to assure strong foundations for future aspirations. Deloitte exemplifies what, how, and why of change so you’re always ready to act ahead. Learn more about [Audit & Assurance Practice](https://www.deloitte.com/global/en/services/audit-assurance/services/assurance.html).

Your work profile

• Design, develop, and deploy Generative AI solutions leveraging Large Language Models (LLMs).
• Build and optimize Retrieval-Augmented Generation (RAG) pipelines for enterprise use cases.
• Develop and manage embedding-based search and retrieval systems using vector databases.
• Create, test, and refine prompts to improve model performance, accuracy, and response quality.
• Integrate and orchestrate LLMs through APIs and AI frameworks for production-grade applications.
• Collaborate with cross-functional teams to identify AI-driven business solutions and use cases.
• Ensure scalability, reliability, and performance of GenAI applications.

Key skills required

• Any Graduate- B.Sc/ B.Tech or in Artificial Intelligence, Data Science, or a related field.
• 8-12 years of experience
• Experience with GenAI frameworks such as LangChain, LlamaIndex, Semantic Kernel, or similar.
• Familiarity with cloud AI platforms such as Azure OpenAI, Azure AI Foundry, AWS Bedrock, or Google Vertex AI.
• Hands-on experience with Retrieval-Augmented Generation (RAG) architectures and frameworks.
• Strong understanding of Embeddings, Vector Stores, and Semantic Search concepts.
• Proven expertise in Prompt Engineering for optimizing LLM outputs.
• Experience working with Large Language Models (LLMs) such as GPT, Claude, Gemini, Llama, Mistral, etc.
• Hands-on experience in LLM initialization, integration, configuration, and deployment.
• Proficiency in using development environments such as VS Code/IntelliJ with GitHub Copilot or equivalent AI-assisted coding tools installed and actively utilized.
• Knowledge of Data Privacy, Security, Access Control, and Compliance within AI solutions.
• Understanding of LLM Cost Optimization techniques, including token management, caching, and model selection strategies.