Deputy Manager | ICC | Mumbai | Data Analytics ICC

Deputy Manager | ICC | Mumbai | Data Analytics ICC • Job requisition ID : 107248 • Location : Mumbai • Entity : Deloitte Touche Tohmatsu India LLP The team Data Technology and Analytics uses innovative tools to analyze vast amounts of data in order to proactively or reactively identify opportunities for revenue enhancement, cost optimization, risk mitigation and strategic advice for our clients. The group specializes in electronic data sourcing, optimization, analysis, interpretation, prediction and modelling. We work as an extension of our Deloitte member firms’ analytics practices and provide dedicated support throughout the project lifecycle. Working on international assignments involving cross-border and cross-service line teams allows you to build your networks across the vast Global Deloitte network. We also have an active international secondment program for our staff after they gain relevant experience, we believe this plays a crucial role in the overall career and personal development of our staff. Learn more about Strategy & Transactions Your work profile As an Assistant Manager focusing on Agentic AI, you will design, build, and deploy autonomous, LLM-powered systems that can reason, plan, and execute tasks across enterprise workflows. You will play a key role in developing agentic architectures, orchestrating multi‑agent systems, and operationalizing GenAI solutions at scale. This role requires strong hands‑on expertise in LLMs, agent frameworks, and AI system design, along with the ability to translate business problems into intelligent, goal‑driven AI agents. In this role, you will: · Collaborate with stakeholders to define use cases suited for Agentic AI and translate them into scalable solution architectures · Implement and configure agentic workflows using LLMs, tools, memory, and orchestration frameworks based on designed architectures and requirements · Build and enhance AI agents capable of multi-step reasoning, planning, and task execution · Develop and deploy RAG-based systems, integrating vector databases, embeddings, and knowledge retrieval pipelines · Support implementation of multi-agent systems for complex problem-solving use cases · Integrate agents with tools, APIs, and enterprise systems to enable task execution · Define and implement evaluation frameworks for agent performance, reasoning quality, and reliability · Ensure solutions are robust, scalable, secure, and aligned with Responsible AI and governance standards · Build and maintain end-to-end pipelines for prompt engineering, agent orchestration, and deployment · Support productionization of AI agents, including monitoring, logging, feedback loops, and continuous improvement · Work with cross-functional teams to ensure AI systems are maintainable, extensible, and aligned with business objectives · Document model logic, assumptions, and validation results in a clear, structured manner. · Clearly communicate agent architectures, behaviours, and limitations to both technical and business stakeholders Key skills required: In order to be considered for the role, your competencies will cover the broad scope of business modelling services, leveraging your professional background and skills such as: Bachelors/ Master’s degree from a reputed institute in Computer Science, AI, Data Science, Information Systems (STEM), Statistics or similar discipline. 3 to 5 years of overall professional IT experience with at least 2+ years in GenAI / LLM-based systems Must have : Strong programming skills in Python Good hands‑on experience with GenAI models, LLMs, and Agentic AI frameworks (OpenAI, Azure OpenAI, Google Gemini, Claude, etc.). Expertise in at least one cloud platform (Azure/AWS/GCP). Strong SQL and data engineering fundamentals. Expertise in building Agentic AI systems, including: Multi-agent orchestration Tool use and action frameworks Planning, reasoning, and memory mechanisms Deep understanding of RAG architectures, including: Vector databases (e.g., Pinecone, FAISS, Azure AI Search) Embeddings and retrieval strategies Prompt engineering and context management Experience with agent frameworks and orchestration tools (e.g., LangChain, LangGraph, Semantic Kernel, AutoGen, CrewAI) Experience deploying AI systems on cloud platforms (Azure/AWS/GCP) Strong understanding of evaluation, guardrails, and safety mechanisms for LLM systems Good to have : Experience building enterprise copilots, chatbots, or AI assistants Familiarity with MLOps / LLMOps tools (MLflow, LangSmith, Weights & Biases, etc.) Knowledge of API integration and microservices architecture Experience with data engineering pipelines supporting AI systems Exposure to UI integration for AI agents (web apps, conversational interfaces) Experience with OpenAI based frameworks Experience with building chatbots Experience with BI tools (Power BI, Tableau). Familiarity with git and DevOps pipelines. This role will also leverage additional skills such as: Strong ability to translate business problems into agent-driven AI solutions Solid understanding of modern AI system design patterns (agentic, retrieval, tool-use) Ability to simplify complex AI concepts for business audiences and influence senior stakeholders. Strong ownership mindset with focus on delivering production-grade AI systems Experience in structuring, presenting, and communicating your work results Proficiency in English and overall communication Effective communication and strong stakeholder management capability with experience engaging cross‑functional and international teams. Flexibility in terms of working hours and working on multiple tools and technologies This profile does not involve extensive travel for work. Hybrid is our default way of working. Each domain has customized the hybrid approach to their unique needs. Knowledge of existing and upcoming platforms and vendors in the market with and leveraging them creatively into wider business solutions This role will also leverage additional skills such as: Ability to create advanced analytical models that are fit for purpose and scalable, to identify insightful trends and patterns from the data Understanding existing and upcoming AI frameworks and realizing them to appropriate assets/solutions Experience in structuring, presenting, and communicating your work results Ability to present complex business and technology problems and solutions in a simple form to address the level of audience from C-level, executive sponsors, and business owners to engineering and operational teams Strong appetite for innovation and emerging technologies Provide guidance on best practices, project execution, and professional development. Demonstrate ability to think strategically about business, create technical solutions with end users in mind, motivate and mobilize resources, as well as deliver results Flexibility in terms of working hours and working on multiple tools and technologies This profile does not involve extensive travel for work. Hybrid is our default way of working. Each domain has customised the hybrid approach to their unique needs.