AI / GenAI Engineering Responsibilities
● Design and develop GenAI / LLM-based components for order orchestration — intelligent document ingestion,
field extraction, enrichment, validation and agentic workflows.
● Build and orchestrate agentic, multi-agent AI pipelines (Observe → Plan → Act → Reflect) using frameworks
such as LangGraph.
● Engineer prompts and reasoning chains for LLMs (Azure OpenAI / GPT) — grounding, structured outputs and
tool / function calling.
● Implement Retrieval-Augmented Generation (RAG) with embeddings and vector / semantic search where
required.
● Integrate document intelligence / OCR (Azure Document Intelligence & Cognitive Services) to read customer
emails and PO documents.
● Design human-in-the-loop (HITL) review flows and guardrails so users stay in control of AI outputs.
● Develop services / APIs (Python / FastAPI) and integrate with SAP (S/4HANA) for Sales Order creation and
enterprise data exchange.
● Implement evaluation, testing and monitoring of AI outputs (accuracy, hallucination / drift checks) and apply
Responsible-AI practices.
● Deploy and support AI services on Azure (AI Foundry, AKS / Kubernetes, Key Vault) with CI/CD.
Technical Skills — AI
Must-have:
● Strong Python for AI / backend; API development (FastAPI / REST).
● Hands-on with LLMs / Azure OpenAI (GPT models): prompt engineering, function / tool calling, structured
outputs.
● Agentic AI / multi-agent orchestration (LangGraph / LangChain or equivalent).
● Document intelligence / OCR (Azure Document Intelligence / Cognitive Services).
● RAG, embeddings and vector databases (e.g., Azure Cosmos DB / vector store).
● Microsoft Azure AI stack (Azure AI Foundry, Cognitive Services).
● Enterprise integration (SAP APIs / REST / JSON).
● Git, containerisation (Docker) and cloud deployment (AKS / Kubernetes, Azure Key Vault).
Good-to-have:
● Front-end: Angular / React (CSR / workbench UI).
● LLMOps / MLOps, model evaluation, guardrails and Responsible AI.
● Vision-language / multimodal models.
● Databricks / data engineering; SQL & NoSQL.
●
Certifications
Required / strongly preferred:
● Microsoft Certified: Azure AI Engineer Associate (AI-102).
● Microsoft Certified: Azure AI Fundamentals (AI-900) or Azure Fundamentals (AZ-900).
Preferred:
● Microsoft Certified: Azure Developer Associate (AZ-204).
● Microsoft Certified: Azure Solutions Architect Expert (AZ-305).
Nice-to-have:
● Certified Kubernetes Application Developer (CKAD) / Administrator (CKA).
● GenAI / LLM certifications (e.g., NVIDIA Generative AI, Google Cloud Generative AI, AWS Certified Machine
Learning, Databricks Generative AI Fundamentals).
● Prompt Engineering / Responsible AI credentials.
● SAP integration-related certification.
Competencies
Competency Description Type
Software Engineering
Processes
Working knowledge and awareness of best practices in an SDLC process to execute tasks at optimal levels; awareness of estimation techniques
and tools.
Professional
Design Knowledge Knows design methodologies, patterns / anti-patterns, tools and
models to translate high-level solutions to detailed design. Professional
Requirements Engineering Thorough knowledge of requirements elicitation and analysis tools and
techniques to define and document requirements. Professional
Configuration & Release Management
Knows best practices and tools in CM and/or release management to
plan and implement CM for a single instance. Professional
Build Working knowledge of a build tool, best practices and algorithms to
increase productivity; unit-testing tools and techniques. Professional
GenAI / LLM Application Engineering
Designs and builds LLM-based solutions — prompting, structured
outputs, tool calling and evaluation. Professional (AI)
Agentic AI & Orchestration Builds multi-agent / agentic workflows and reasoning pipelines (e.g.,
LangGraph). Professional (AI)
RAG & Applied ML Retrieval-Augmented Generation, embeddings, vector search and
document intelligence. Professional (AI)
Cloud (Azure) AI Services Uses Azure AI Foundry, Cognitive Services and deploys on AKS with
secure config. Professional (AI)
Responsible AI Awareness Applies guardrails, evaluation and governance for safe, reliable AI
outputs. Professional (AI)
Analytical Ability Integrates data from multiple sources to build cause-effect linkages and
evaluate solution alternatives. Behavioural
High Impact Communication Structures thoughts, listens, communicates with conviction in simple
language and shares views confidently. Behavioural
Planning & Organizing Prioritises tasks by criticality and interdependency, and plans
proactively using historical data. Behavioural
Educational Qualification & Experience
Educational Qualification Engineering Graduate (BE / B.Tech / MCA or equivalent).
Minimum Experience 3+ years overall.
AI / IOOP Experience (added) Hands-on experience building AI / GenAI / LLM applications — typically 1+ year on GenAI /
agentic solutions, with API and cloud (Azure) integration.