Manager - Python

We're looking for an Automation Specialist with 4-5 years of experience to join the Operational Excellence team. In this role, you will design, develop, and implement automation solutions using Python, Power Platform, and increasingly, AI/LLM-driven workflows. You'll collaborate with cross-functional teams to understand business requirements and deliver scalable solutions — including intelligent, agentic automations — that help the organization operate more efficiently. Location – Mumbai

Responsibilities

Understand business requirements and translate them into automated flows/scripts, including AI-augmented workflows where LLMs add reasoning, summarization, or decision-making capability; Create Power Automate flows based on various triggers; Integrate with other apps (model-driven, canvas apps); Create and trigger Adaptive Cards via MS Teams/Outlook using Power Automate & Power BI APIs; Record/retrieve data in a DB/SharePoint List using PowerApps & Power Automate; Work with multiple data formats — establish connections with SharePoint, REST APIs, DBs, Dataflows, and vector/embedding stores; Design and implement RAG (Retrieval-Augmented Generation) pipelines — document ingestion, chunking, embedding generation, vector search, and grounded response generation; Integrate with LLM APIs (OpenAI, Anthropic Claude, Azure OpenAI, etc.) for tasks like summarization, classification, extraction, and conversational interfaces; Design and build agentic workflows — multi-step, tool-using automations where an LLM plans and orchestrates actions (e.g., using frameworks like LangChain, LangGraph, Semantic Kernel, or custom orchestration in Python); Understand Power BI dataflows and identify data limitations, proposing solutions accordingly; Build applications/flows using Python to automate processes, including AI-driven ones; Coordinate with development teams to build RESTful web services, including endpoints that wrap or serve LLM-based functionality; Write unit test cases for developed code, including evaluation/validation approaches for LLM-based outputs (accuracy, hallucination checks, groundedness); Participate in UAT; diagnose and troubleshoot bugs, integration issues, and AI-output quality issues; Write reusable, effective, and scalable Python code, including for prompt engineering, chaining, and orchestration logic; Architect and build applications, coordinating with teams responsible for other layers of the product/AI infrastructure; Write well-documented, clean code — including documentation of prompts, chains, and agent logic for maintainability

Requirements

Bachelor's degree in computer science, information technology, or a related field; 4-5 years of experience as an Automation Consultant or in a similar role; Expert-level proficiency in Python; Hands-on experience with LLM APIs (OpenAI, Anthropic, Azure OpenAI, or similar) and prompt engineering fundamentals; Practical understanding of RAG architectures — embeddings, vector databases (e.g., Pinecone, FAISS, Chroma, Azure AI Search), and retrieval strategies; Familiarity with common databases (RDBMS such as MySQL & NoSQL such as MongoDB); Knowledge of object-oriented programming techniques; Ability to develop functional databases, applications, and servers to support back-end services; Understanding of performance optimization and multithreading concepts (relevant to async LLM calls and batch processing); Strong problem-solving skills and familiarity with code versioning tools such as GitHub; Understanding of REST concepts and building/interacting with REST APIs; SQL; Strong knowledge of Power Automate and creating PowerApps; Knowledge and understanding of Power BI APIs, Dataflows & Power Query Editor; Excellent communication and interpersonal skills; Ability to work independently and as part of a team; Strong attention to detail

Nice to have

Agentic framework experience (LangChain/LangGraph/AutoGen); vector databases; Docker; Exposure to building or working with agentic frameworks (LangChain, LangGraph, AutoGen, Semantic Kernel, or custom agent orchestration); Familiarity with vector stores; Background in containerization tools like Docker