Accenture is seeking an AI / ML Engineer in Pune to develop production-ready applications and systems using AI tools, cloud AI services and generative AI models. The role focuses on integrating, deploying and optimizing cloud-based AI models, with Large Language Models (LLMs) as the required skill.
Responsibilities
Design and implement high-quality application pipelines in cloud or on-premises environments. Develop solutions that may use deep learning, neural networks, chatbots and image processing.
Serve as a subject-matter expert, collaborate with and manage the team, and take responsibility for team decisions. Work across teams on key decisions and solve problems for the immediate team and other teams.
Mentor junior colleagues and continuously improve processes and workflows. Evaluate model outputs and optimize solutions.
Requirements and technical experience:
At least 5 years of experience are required; the posting also specifies a minimum of 5 years of experience with LLMs. A 15-year full-time education is required.
Experience with Java, Python and AI application development, particularly cloud-based model integration, deployment and optimization. Relevant frameworks include Spring AI, LangChain, LangGraph, LlamaIndex, Agno, Pydantic and TensorFlow.
Hands-on experience with Azure Cloud Services for AI, including Azure OpenAI and Azure AI Search, to develop and deploy AI applications at scale.
Proficiency in advanced AI architecture patterns, including Retrieval-Augmented Generation (RAG), Agentic RAG, MCP, Functional Calling and A2A, especially in Azure. Experience with prompt and context engineering, vector databases, embedding and chunking strategies, and real-time data integration.
For full details about this role, click “Apply on company site” to view the employer’s original posting.
Skills for this role
Large Language Models (LLMs)PythonJavaSpring AILangChainLangGraphLlamaIndexAgnoPydanticTensorFlowAzure Cloud ServicesAzure OpenAIAzure AI SearchRetrieval-Augmented Generation (RAG)Agentic RAGMCPFunctional CallingA2APrompt EngineeringContext EngineeringVector DatabasesEmbedding StrategiesChunking StrategiesReal-Time Data IntegrationModel EvaluationModel OptimizationCloud AI Model IntegrationAI Application DeploymentGenerative AIDeep Learning
Your skill match
Checking your profile…
YOUR NEXT STEP
Get interview-ready for this role
A focused preparation guide, built around this job’s responsibilities and requirements.
✦ AI-generated guide
Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.
01
Understand the role
What matters most
Production-Ready AI Pipeline Development
The role requires designing and implementing robust application pipelines in cloud or on-premises environments. You must demonstrate the ability to move beyond experimental prototypes to scalable, high-quality systems using Azure Cloud Services and frameworks like LangChain or Spring AI.
Advanced LLM Architecture Implementation
You are expected to be a subject matter expert in advanced patterns such as Retrieval-Augmented Generation (RAG), Agentic RAG, and functional calling. Success depends on your ability to optimize embedding strategies, manage vector databases, and ensure real-time data integration within an Azure ecosystem.
Technical Leadership and Mentorship
Beyond individual coding, you must manage team performance, make critical architectural decisions, and mentor junior colleagues. You will be responsible for evaluating model outputs and continuously refining workflows to improve overall team delivery.
Likely challenges to think through
Balancing Rapid Innovation with Production Stability
The fast-evolving nature of GenAI frameworks (e.g., LangGraph, Agno) creates a challenge in maintaining stable, production-ready code. Prepare by discussing how you implement rigorous testing and evaluation protocols for non-deterministic LLM outputs to ensure reliability in enterprise environments.
Cross-Team Technical Influence
As an SME working across multiple teams, you will face the challenge of aligning diverse stakeholders on complex architectural decisions. Prepare by reflecting on instances where you had to justify technical trade-offs, such as choosing specific chunking strategies or model architectures, to non-technical or cross-functional partners.
03
Practice your answers
Technical practice
RAG and Agentic RAG Implementation
What to coverExplain the architectural differences between standard RAG and Agentic RAG. Discuss how you handle complex queries that require multi-step reasoning or tool usage within an Azure environment.
Azure AI Search and Vector Database Optimization
What to coverDescribe your approach to optimizing vector search performance in Azure AI Search. Focus on how you select embedding models and chunking strategies to balance retrieval accuracy with latency.
Prompt and Context Engineering
What to coverHow do you manage prompt versioning and context window limitations in production LLM applications? Suggestion: Prepare to discuss techniques for dynamic context injection and prompt template management using frameworks like LangChain or Pydantic.
Model Evaluation and Optimization
What to coverWhat metrics do you use to evaluate the output quality of generative AI models in a production pipeline? Explain how you perform iterative optimization based on these evaluation results.
Integration of AI Frameworks
What to coverCompare the use cases for Spring AI versus LangChain when building enterprise-grade Java or Python applications. Discuss how you integrate these frameworks with Azure OpenAI services.
Real-time Data Integration
What to coverDescribe the challenges of integrating real-time data into an LLM-based system. Focus on maintaining data freshness and consistency while ensuring low-latency responses for the end user.
Functional Calling and A2A Patterns
What to coverExplain how you implement functional calling to allow LLMs to interact with external APIs or internal tools. Discuss the security and reliability considerations when designing Agent-to-Agent (A2A) communication.
Production Pipeline Design
What to coverWhat are the key components of a production-ready AI pipeline? Suggestion: Address CI/CD for models, monitoring for drift, and the deployment of scalable infrastructure on Azure.
Behavioural practice
Mentoring and Team Development
What to coverDescribe a situation where you had to mentor a junior colleague on complex AI concepts. Focus on how you structured your guidance and the impact it had on their technical growth.
Cross-functional Decision Making
What to coverProvide an example of a time you had to make a key technical decision that affected multiple teams. Explain how you gathered input, managed conflicting priorities, and ensured team alignment.
Process Improvement
What to coverTell me about a time you identified an inefficiency in an existing development workflow. Detail the steps you took to improve the process and how you measured the success of those changes.
Handling Technical Disagreements
What to coverDescribe a scenario where you disagreed with a team member or stakeholder regarding a technical solution. How did you resolve the conflict while maintaining a productive working relationship?
02
Sharpen your knowledge
Topics to prepare
Advanced RAG Architectures
Review the conceptual differences between standard RAG and Agentic RAG, including the role of MCP and functional calling. Exercise: Design a system architecture diagram for an Azure-based RAG pipeline that incorporates agentic decision-making to handle complex multi-step user queries.
Production-Ready AI Pipelines
Focus on the lifecycle of deploying models in cloud environments, specifically addressing model evaluation and optimization. Exercise: Outline a strategy for monitoring and evaluating LLM output quality in a production environment, focusing on latency, accuracy, and cost optimization on Azure.
Skills to revise
Azure AI Services Integration
Hands-on experience with Azure OpenAI and Azure AI Search is required. Exercise: Create a prototype application using Python or Java that integrates Azure AI Search as a vector store to perform retrieval-augmented generation for a specific document set.
LLM Orchestration Frameworks
Proficiency in frameworks like LangChain, LangGraph, or LlamaIndex is essential. Exercise: Build a small workflow using LangGraph or LangChain that demonstrates a multi-agent interaction or a complex chain of thought process to solve a specific data processing task.
Prompt and Context Engineering
Mastering the input side of LLMs is critical for high-quality outputs. Exercise: Develop a set of system prompts and context injection strategies for a chatbot, testing how different chunking strategies and embedding techniques affect the relevance of the retrieved information.
04
Your preparation checklist
Architect a RAG-based solution on Azure
Prepare a technical walkthrough of an end-to-end RAG pipeline you have built. Focus on your implementation of Azure OpenAI and Azure AI Search, specifically detailing your chunking strategies, embedding models, and how you handled real-time data integration.
Demonstrate proficiency in Agentic workflows
Prepare to discuss your experience with Agentic RAG and functional calling. Be ready to explain how you utilized frameworks like LangChain, LangGraph, or LlamaIndex to build autonomous agents that perform multi-step reasoning or tool-use tasks.
Prepare examples of model optimization and evaluation
Compile specific examples where you evaluated model outputs and optimized performance. Be prepared to discuss how you managed prompt engineering and context engineering to improve accuracy, reduce latency, or lower costs in a production environment.
Develop a leadership and mentorship narrative
Since the role requires SME-level responsibilities, prepare concrete examples of how you have mentored junior engineers, managed team technical decisions, and improved development workflows or CI/CD processes for AI applications.
Formulate strategic questions for the interviewers
Prepare questions regarding the team's current approach to scaling AI applications on Azure, how they handle model governance, and the specific challenges they face when integrating generative AI into existing enterprise systems.
Availability last checked: 4 Nov 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy
Share this opportunityKnow someone who would be a great fit?