New opportunity

AI / ML Engineer

Accenture · Bengaluru, India

About this role

Accenture is hiring an AI / ML Engineer in Bengaluru to own NLP and generative AI solutions from design through production. The role focuses on scalable, cloud-native applications powered by large language models on Azure, with production-ready cloud or on-premises application pipelines.

Responsibilities

Develop LLM-based chatbots, retrieval-augmented generation (RAG) systems and AI assistants; design data ingestion, preprocessing and document-update pipelines.
Implement prompt engineering, embeddings, vector search and evaluation. Build asynchronous REST APIs with FastAPI or Flask, and manage LLM API rate limits, retries, load balancing and fault tolerance.
Deploy, monitor and scale applications on Azure. Deliver proofs of concept and minimum viable products under tight timelines, mentor junior colleagues and support team delivery.

Requirements

Large Language Models are the stated required skill. The posting also calls for advanced Python; NLP and generative AI experience, including RAG and prompt engineering; Azure OpenAI or other LLM platforms; asynchronous FastAPI or Flask; Azure App Service, Storage, AI Search and Cosmos DB; and CI/CD, logging and monitoring using tools such as Jenkins and OpenTelemetry.
A minimum of 7.5 years of experience and 15 years of full-time education are required. The experience section separately specifies 5+ years in NLP, AI or ML and proven production deployment of generative AI systems; the listing header states an experience range of 5–10 years.
Other technologies and areas listed include vector databases, agentic libraries such as LangChain, LangGraph and AutoGen, computer vision and document processing.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonNatural Language ProcessingGenerative AILarge Language ModelsRetrieval-Augmented GenerationPrompt EngineeringEmbeddingsVector SearchVector DatabasesAzure OpenAIAzure CloudAzure App ServiceAzure StorageAzure AI SearchAzure Cosmos DBFastAPIFlaskREST APIsCI/CDJenkinsOpenTelemetryLoggingMonitoringLangChainLangGraphAutoGenComputer VisionDocument ProcessingData IngestionData Preprocessing

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

End-to-End Generative AI Lifecycle Management
The role requires owning solutions from design through production. You must demonstrate proficiency in building RAG systems, managing data ingestion pipelines, and ensuring production-ready quality for LLM-based applications on Azure.
Scalable Cloud-Native Architecture
You are expected to build and deploy applications using Azure services like App Service, AI Search, and Cosmos DB. This includes implementing asynchronous REST APIs with FastAPI or Flask and managing complex production concerns like rate limiting, fault tolerance, and load balancing.
Technical Leadership and Delivery
Beyond individual development, you must deliver rapid PoCs and MVPs under tight timelines while mentoring junior colleagues. You should be prepared to discuss how you balance high-velocity development with maintaining robust CI/CD and monitoring standards.

Likely challenges to think through

Balancing Rapid Prototyping with Production Stability
The requirement to deliver MVPs under tight timelines while ensuring production-ready reliability creates a trade-off between speed and technical debt. Prepare by discussing your strategies for building modular, testable code that can evolve from a PoC to a stable, monitored production system.
Managing LLM Non-Determinism in Production
Implementing RAG and agentic workflows requires handling the inherent unpredictability of LLMs. You should be prepared to explain how you implement evaluation frameworks, logging, and observability using tools like OpenTelemetry to maintain performance and accuracy in live environments.

Practice your answers

Technical practice

Designing RAG Pipelines for Production

What to coverExplain your approach to building a robust retrieval-augmented generation pipeline. Focus on data ingestion, preprocessing, and the trade-offs between different embedding strategies and vector search configurations in Azure AI Search.

Managing LLM API Reliability

What to coverHow do you handle rate limits, retries, and fault tolerance when integrating LLM APIs into a production application? Discuss strategies for implementing load balancing and circuit breakers within an asynchronous FastAPI or Flask environment.

Prompt Engineering and Evaluation

What to coverDescribe your methodology for iterative prompt engineering and systematic evaluation of LLM outputs. Suggestion: Prepare to discuss how you measure accuracy, hallucination rates, and performance consistency in a production setting.

Azure Cloud-Native Deployment

What to coverDetail your experience deploying and scaling AI applications on Azure. Specifically, address how you utilize Azure App Service, Cosmos DB, and Storage to ensure high availability and performance for LLM-based services.

Agentic Workflows with LangChain or AutoGen

What to coverDiscuss the implementation of agentic workflows using libraries like LangChain, LangGraph, or AutoGen. Explain how you manage state, tool selection, and multi-agent coordination in complex AI assistants.

Monitoring and Observability for AI

What to coverHow do you implement logging and monitoring for generative AI systems using tools like OpenTelemetry? Explain how you track model performance, latency, and drift in a production environment.

CI/CD for Machine Learning

What to coverDescribe your process for integrating CI/CD pipelines for AI models using Jenkins. Focus on how you automate testing, versioning, and deployment to ensure production-ready quality.

Document Processing and Computer Vision

What to coverHow do you integrate document processing or computer vision tasks into a broader generative AI solution? Discuss the challenges of extracting structured data from unstructured documents for downstream LLM consumption.

Behavioural practice

Delivering Under Tight Timelines

What to coverDescribe a situation where you had to deliver a Proof of Concept or MVP under significant time pressure. Focus on how you prioritized features, managed stakeholder expectations, and ensured the quality of the final deliverable.

Mentoring and Team Support

What to coverProvide an example of how you have mentored junior colleagues or supported team delivery in a technical project. Highlight your approach to knowledge sharing and how you helped others overcome technical hurdles.

Handling Technical Trade-offs

What to coverTell me about a time you had to make a difficult technical trade-off between model performance and production constraints (e.g., latency vs. accuracy). Explain your decision-making process and the resulting impact.

Navigating Production Incidents

What to coverDescribe a challenging production issue you encountered with an AI system. Explain the steps you took to diagnose the root cause, the actions taken to resolve it, and how you prevented future occurrences.

Sharpen your knowledge

Topics to prepare

Retrieval-Augmented Generation (RAG) Architecture
Review the end-to-end flow of RAG systems, specifically focusing on data ingestion, document preprocessing, and vector search integration. Exercise: Design a conceptual pipeline for a document-heavy chatbot, detailing how you would handle chunking strategies and metadata filtering to improve retrieval accuracy.
Production-Grade LLM Application Resilience
Understand strategies for managing LLM API constraints in production environments. Exercise: Draft a technical approach for implementing fault tolerance, including exponential backoff for retries, load balancing across multiple model endpoints, and rate-limiting mechanisms for high-traffic asynchronous APIs.
Agentic Workflow Orchestration
Study the integration of agentic libraries like LangChain, LangGraph, or AutoGen for complex task automation. Exercise: Map out a multi-agent workflow where one agent handles document retrieval and another performs synthesis, identifying how state management is handled between steps.

Skills to revise

Asynchronous Python API Development
Develop proficiency in building non-blocking REST APIs using FastAPI. Exercise: Create a sample FastAPI endpoint that performs an asynchronous call to an LLM, incorporating logging and OpenTelemetry tracing to monitor request latency and success rates.
Azure Cloud AI Services Implementation
Gain hands-on familiarity with the Azure ecosystem for AI deployment. Exercise: Configure a mock deployment pipeline using Azure App Service and Cosmos DB, ensuring that you can demonstrate how to store and query vector embeddings using Azure AI Search.
Prompt Engineering and Evaluation
Master the iterative process of prompt refinement and systematic evaluation. Exercise: Develop a test suite for a specific prompt template, defining metrics for relevance and hallucination, and document how you would automate this evaluation within a CI/CD pipeline using Jenkins.

Your preparation checklist

Prepare a production-grade RAG architecture walkthrough
Draft a technical summary of a past project where you deployed a RAG system to production. Focus on your design choices for data ingestion pipelines, vector database selection (e.g., Azure AI Search), and how you handled asynchronous API calls, rate limiting, and fault tolerance using FastAPI.
Demonstrate expertise in LLM orchestration and agentic workflows
Prepare to discuss your experience with agentic libraries like LangChain, LangGraph, or AutoGen. Be ready to explain how you implemented specific agentic patterns to solve complex tasks and how you managed prompt engineering and evaluation metrics in a production environment.
Map your Azure experience to the required stack
Create a summary table matching your experience with Azure App Service, Cosmos DB, and Azure OpenAI to the specific requirements of this role. Highlight instances where you configured CI/CD pipelines (e.g., Jenkins) and implemented observability using OpenTelemetry for monitoring LLM performance.
Develop targeted questions for the interviewers
Prepare questions regarding the team's current approach to LLM evaluation, how they handle data privacy and security in document processing pipelines, and the specific challenges they face when scaling generative AI solutions on Azure.

Availability last checked: 4 Nov 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…