New opportunity

Gen AI Engineer

Infosys Limited · Bangalore, India

About this role

Infosys Limited seeks a Gen AI Engineer in Bangalore for its Cloud & Infrastructure Services service line. The posting specifies 5–9 years of work experience.

Responsibilities

Design, develop and deploy AI-powered automation solutions using Python and generative AI.
Build and improve applications using large language models, retrieval-augmented generation and prompt engineering; integrate solutions with enterprise systems, workflows and APIs.
Support model development, testing, deployment, monitoring and optimization. Troubleshoot deployed AI applications and automation workflows.
Work with cross-functional teams to identify and implement AI use cases, follow Responsible AI, security and governance guidelines, and contribute reusable components, automation assets and internal knowledge.

Education and preferred skills:

Listed educational requirements are Bachelor of Engineering, BTech, Bachelor of Science or Master of Engineering.
Preferred skills are Python, artificial intelligence, retrieval-augmented generation and LLMOps.

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

Skills for this role

PythonGenerative AILarge Language ModelsRetrieval-Augmented GenerationPrompt EngineeringLLMOpsAI automationAPI integrationModel deploymentModel monitoringModel optimizationResponsible AISecurity governance

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

Deliver AI-powered automation
The posting calls for designing, developing and deploying automation solutions using Python and generative AI. Prepare to explain the engineering decisions behind a solution from design through deployment.
Build useful LLM applications
Building and enhancing applications with LLMs, retrieval-augmented generation (RAG) and prompt engineering is an explicit responsibility. Prepare to explain when retrieval or prompt changes improve an application.
Connect AI to enterprise workflows
The role integrates AI solutions with enterprise systems, workflows and APIs. Prepare to discuss how an integration handles inputs, outputs and failures.
Own the deployed lifecycle
The posting includes model development, testing, deployment, monitoring, optimization and troubleshooting of deployed applications and automation workflows. Prepare to describe how test results and operational signals guide fixes.
Work within governance and share reusable work
The role involves identifying use cases with cross-functional teams, following Responsible AI, security and governance guidelines, and contributing reusable components and internal knowledge.

Likely challenges to think through

Balancing answer quality with complexity
Preparation hypothesis, not a reported employer problem: RAG and prompt changes may improve answers while adding retrieval dependencies and failure points. Prepare to compare approaches using representative tests, including cases where retrieved material is missing or unhelpful.
Integrating without weakening controls
Preparation hypothesis, not a reported employer problem: connecting an AI application to enterprise APIs may make automation useful while increasing exposure to sensitive data or unintended actions. Prepare to explain how you would limit data access and check outputs against security and governance guidelines.
Improving a live solution without disrupting it
Preparation hypothesis, not a reported employer problem: optimization can change application behavior while deployed workflows still need to work reliably. Prepare to describe how monitoring, regression tests and troubleshooting evidence would inform a change and its release.

Practice your answers

Technical practice

How would you design a Python-based AI automation solution from development through deployment?

What to coverThe posting calls for designing, developing and deploying AI-powered automation using Python and generative AI. Suggested preparation: Explain how you would structure the application, handle failures and decide which steps need human review.

How would you build and improve a retrieval-augmented generation application?

What to coverThe posting explicitly mentions RAG and LLM applications. Suggested preparation: Walk through retrieving relevant information, supplying it to the model and checking whether answers are grounded in that information.

How would you improve a prompt that produces inconsistent results?

What to coverPrompt engineering is a stated responsibility. Suggested preparation: Discuss clear instructions, representative test cases and the trade-off between adding context and keeping prompts manageable.

How would you integrate an AI application with an enterprise workflow and API?

What to coverThe posting calls for integration with enterprise systems, workflows and APIs. Suggested preparation: Explain how you would define inputs and outputs, handle API failures and prevent an unreliable model response from disrupting the workflow.

How would you test an LLM application before deployment?

What to coverThe role includes model testing and application development. Suggested preparation: Describe test cases for expected outputs, unsupported answers and workflow failures, and explain how you would evaluate changes against a baseline.

What would you check when deploying an updated AI application?

What to coverDeployment is part of the advertised work, and LLMOps is a preferred skill. Suggested preparation: Discuss versioning, pre-release checks and how you would detect and respond to a regression after release.

How would you monitor and optimize a deployed LLM application?

What to coverThe posting explicitly includes monitoring and optimization. Suggested preparation: Distinguish output quality from operational measures such as latency and failures, then explain how you would weigh improvements against resource use.

How would you diagnose a deployed AI workflow that has started returning poor results?

What to coverTroubleshooting deployed AI applications and automation workflows is a stated duty. Suggested preparation: Describe how you would isolate retrieval, prompts, model behavior and system integration as possible causes before choosing a fix.

How would you apply Responsible AI, security and governance guidelines to an AI solution?

What to coverThe posting requires following Responsible AI, security and governance guidelines. Suggested preparation: Explain how you would review data access, sensitive outputs and the need for human oversight without assuming a particular company policy.

Behavioural practice

Tell me about a time you worked across teams to select an AI or automation use case.

What to coverThe posting calls for collaborating with cross-functional teams to identify and implement AI-driven use cases. Suggested preparation: Use a real situation/action/result example showing how you understood the need, agreed on a feasible scope and measured the outcome.

Describe a time you resolved an issue in a deployed application or automation workflow.

What to coverTroubleshooting deployed solutions is part of the role. Suggested preparation: Use a real situation/action/result example that separates your diagnosis and actions from the eventual result, including how you communicated with affected people.

Tell me about a time a security or responsible-use concern changed your technical approach.

What to coverThe posting requires following Responsible AI, security and governance guidelines. Suggested preparation: Use a real situation/action/result example explaining the concern, the trade-off you made and the outcome; do not assume a specific Infosys policy.

Describe a time you made a component reusable or shared knowledge from a project.

What to coverThe role includes contributing reusable components, automation assets and internal knowledge. Suggested preparation: Use a real situation/action/result example covering what you made easier for others to use, how you documented it and what changed as a result.

Sharpen your knowledge

Topics to prepare

LLMs, RAG and prompt engineering
The role explicitly calls for applications using LLMs, retrieval-augmented generation (RAG) and prompt engineering. As a revision exercise, sketch how a question moves through retrieval, prompt construction and answer generation; explain where an unsupported answer could arise.
AI application lifecycle and LLMOps
The posting names model development, testing, deployment, monitoring and optimization, and lists LLMOps as a preferred skill. As a revision exercise, map these stages for a deployed AI application and identify what evidence would prompt an update or rollback.
Enterprise integration design
The role involves integrating AI solutions with enterprise systems, workflows and APIs. As a revision exercise, diagram an AI-powered workflow and explain its inputs, API boundaries, failure handling and outputs.
Responsible AI, security and governance
The posting requires following Responsible AI, security and governance guidelines. As a revision exercise, review a hypothetical AI workflow for sensitive inputs, inappropriate outputs and access risks, then propose checks; these examples are preparation suggestions, not stated company controls.

Skills to revise

Python automation implementation
Python is a preferred skill and is named for designing and deploying AI-powered automation. For practice, write a small Python workflow that validates an input, invokes a simulated AI step and handles an error.
RAG application prototyping
Building and enhancing RAG applications is an explicit responsibility. For practice, assemble a small document-based question-answering flow, compare answers with and without retrieved context, and note unsupported claims.
Prompt iteration
Prompt engineering is explicitly named. For practice, revise a prompt against a few representative inputs and record which changes improve answer relevance or adherence to the requested output.
API and workflow integration
Integration with APIs and enterprise workflows is stated. For practice, connect a Python AI workflow to a sample API interface and test malformed responses and unavailable-service behavior; the sample interface is a practice suggestion.
Testing and troubleshooting deployed behavior
The role includes testing, monitoring, optimization and resolving issues in deployed AI applications. For practice, define test cases for answer quality and workflow failures, inspect sample failure records, and explain a targeted fix and how you would verify it.

Your preparation checklist

Map the posted criteria to evidence
Prepare a one-page list of examples relevant to the stated 5–9 years of experience, listed degrees, and preferred Python, AI, RAG, and LLMOps skills. Note any areas you would need to discuss rather than assuming a match.
Build a Python automation walkthrough
Prepare a project narrative or illustrative design showing the use case, your design and development decisions, deployment approach, and outcome. The posting calls for AI-powered automation and integration with enterprise workflows or APIs.
Practice a RAG application design
Sketch how an LLM application would retrieve information and use prompts to produce an answer. Explain your retrieval and prompt choices, plus how you would test response quality; this is a preparation exercise, not a stated interview task.
Document an operational troubleshooting example
Prepare a concise account of a deployed AI application issue: how you detected it, investigated it, resolved it, and monitored the result. Connect it to the posting’s testing, deployment, monitoring, optimization, and troubleshooting responsibilities.
Prepare a responsible-AI review
Take one proposed AI automation use case and write down security, governance, and responsible-AI risks you would check before deployment. The posting requires following guidelines but does not specify a particular framework.
Show collaboration and reuse
Prepare an example of working with other functions to select or implement a use case, and identify a component, automation asset, or knowledge note that could be reused. Keep the example limited to work you can substantiate.
Draft questions for the employer
Ask which Cloud & Infrastructure Services use cases are priorities, how AI solutions connect to existing systems, and what monitoring and governance expectations apply to this role.

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…