New opportunity

Gen AI Developer

INFOSYS LIMITED · Bangalore, India

About this role

Infosys Limited is seeking a Gen AI Developer in Bangalore for its Topaz COE unit. The role focuses on building scalable Python applications that deploy generative AI solutions and integrate AI models into domain-specific applications. The posting specifies 2–6 years of work experience and a Bachelor of Engineering degree.

Responsibilities and requirements:

Develop modular Python applications using FastAPI and asyncio, with attention to parallel processing.
Build and manage cloud infrastructure using AWS S3, Lambda, DynamoDB, ECS and EKS; automate infrastructure provisioning and configuration with Terraform.
Work with data scientists, ML engineers and product teams to integrate AI models into applications.
Ensure GenAI systems meet production requirements for scalability, reliability and security. Monitor and optimize performance using AWS CloudWatch.
Participate in code reviews and documentation, follow Python development best practices, and keep current with developments in GenAI, cloud computing, MLOps and DevOps.

Preferred skills: Generative AI for Data Analytics, prompt engineering and Databricks Machine Learning.

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

Skills for this role

PythonFastAPIasyncioModular application designParallel processingGenerative AIAWSAmazon S3AWS LambdaAmazon DynamoDBAmazon ECSAmazon EKSTerraformAmazon CloudWatchMLOpsDevOpsPrompt EngineeringDatabricks Machine LearningGenerative AI for Data AnalyticsCode reviewTechnical documentation

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

Build deployable GenAI applications
The role calls for scalable, modular Python applications using FastAPI, asyncio and parallel processing to deploy generative AI solutions.
Own AWS infrastructure and automation
The posting names S3, Lambda, DynamoDB, ECS and EKS for cloud infrastructure, with Terraform for provisioning and configuration.
Integrate models into domain applications
Work with data scientists, ML engineers and product teams to incorporate AI models into domain-specific applications.
Meet production standards
Ensure GenAI systems are scalable, reliable and secure, then monitor and optimize performance using tools such as AWS CloudWatch.
Maintain engineering quality
Contribute to code reviews, documentation and Python best practices. Generative AI for data analytics, prompt engineering and Databricks Machine Learning are listed as preferred skills, not core duties.

Likely challenges to think through

Concurrency versus predictable performance
Preparation hypothesis: asyncio and parallel processing can improve throughput while making resource use and failures harder to control. Prepare to explain how you would bound concurrent work, test under load and diagnose latency.
Model integration versus application needs
Preparation hypothesis: a model may work in isolation but need adaptation for a domain-specific application. Prepare an example of defining an interface and validating outputs with data science, ML engineering and product partners.
Deployment speed versus operational safeguards
Preparation hypothesis: automating deployment across AWS services can accelerate changes while increasing the need for reliable configuration and security checks. Prepare to discuss Terraform review, failure recovery and CloudWatch signals without assuming a particular Infosys process.

Practice your answers

Technical practice

How would you structure a modular FastAPI application that serves a generative AI model?

What to coverThe posting calls for scalable, modular Python applications using FastAPI. Suggested focus: explain how you would separate API handling, model integration and application logic, and how you would test those boundaries.

When would you use asyncio versus parallel processing in a Python AI application?

What to coverThe posting names both asyncio and parallel processing. Suggested focus: distinguish concurrent I/O from CPU-intensive work, and discuss how each choice affects throughput and resource use.

How would you integrate an AI model into a domain-specific application?

What to coverThe role involves working with data scientists, ML engineers and product teams on model integration. Suggested focus: discuss input and output contracts, domain validation, failure handling and how you would agree on measurable application behaviour.

How would you choose among the AWS services named for this role?

What to coverThe posting names S3, Lambda, DynamoDB, ECS and EKS for building and managing cloud infrastructure. Suggested focus: compare their roles in a possible deployment, including operational complexity, scaling needs and data access, without assuming one prescribed architecture.

How would you manage infrastructure changes with Terraform?

What to coverThe posting specifies Terraform for provisioning and configuration. Suggested focus: explain reusable configuration, reviewing proposed changes, managing environment differences and reducing deployment risk.

How would you test whether a GenAI service is ready to scale reliably?

What to coverThe posting requires production-grade scalability and reliability. Suggested focus: describe load and failure testing, useful service metrics, and trade-offs among latency, capacity and recovery behaviour.

What security risks would you consider when deploying a GenAI application?

What to coverProduction-grade security is an explicit responsibility. Suggested focus: discuss access to models and data, handling sensitive inputs and outputs, and how you would verify that safeguards work.

How would you investigate a sudden increase in AI-service latency?

What to coverThe posting calls for monitoring and performance optimization using tools such as AWS CloudWatch. Suggested focus: explain which metrics or logs you would inspect, how you would isolate the bottleneck and how you would check that a change improved performance.

How would you evaluate a prompt change before releasing it?

What to coverPrompt engineering is listed as a preferred skill. Suggested focus: compare outputs against representative tasks, check consistency and errors, and weigh quality against latency or cost.

How might you assess a generative AI use case for data analytics?

What to coverGenerative AI for Data Analytics and Databricks Machine Learning are preferred skills, not stated requirements for every project. Suggested focus: discuss data quality, evaluation against the intended analytical task and, if you have relevant experience, where Databricks Machine Learning would fit.

Behavioural practice

Tell me about a time you coordinated with specialists to integrate a model or service into an application.

What to coverThe posting calls for collaboration with data scientists, ML engineers and product teams. Suggested focus: use a real situation, explain how you agreed on responsibilities and interfaces, and state the result.

Describe a time you diagnosed and improved a production performance or reliability issue.

What to coverMonitoring, optimization and reliability are stated responsibilities. Suggested focus: give the situation, the evidence you used, the actions you took and the measured result, if available.

Tell me about a code review that improved maintainability or prevented a problem.

What to coverThe role includes code reviews, documentation and Python best practices. Suggested focus: describe the situation, how you gave or used feedback, what changed and the result.

Describe a time you had to balance delivery speed with security or reliability.

What to coverThe posting expects production-grade security and reliability alongside application delivery. Suggested focus: explain the competing needs, your actions and trade-offs, and the outcome without claiming every risk was eliminated.

Tell me about a time you applied a new technical development to your work.

What to coverThe posting asks the developer to stay current with GenAI, cloud computing, MLOps and DevOps. Suggested focus: give a real situation, explain how you evaluated and applied the idea, and describe the result.

Sharpen your knowledge

Topics to prepare

Python concurrency and modular design
The posting calls for FastAPI, asyncio, modular applications and parallel processing. Suggested revision exercise: explain when concurrent requests benefit from asyncio, when work needs parallel processing, and how you would separate API, model-integration and infrastructure concerns.
AWS service roles
S3, Lambda, DynamoDB, ECS and EKS are explicitly named. Suggested revision exercise: sketch a GenAI application and explain what each service could do, including why you might choose Lambda rather than an ECS or EKS deployment for a particular component.
Production GenAI design
The role requires scalability, reliability and security when deploying GenAI solutions. Suggested revision exercise: walk through a model-backed request and identify likely capacity, failure and security concerns, plus what you would monitor. The specific mitigations are preparation suggestions, not prescribed by the posting.
Model integration across teams
The posting describes working with data scientists, ML engineers and product teams to integrate models into domain-specific applications. Suggested revision exercise: outline the interface, input and output expectations, and handoffs you would clarify before integrating a model.
Preferred GenAI specializations
Generative AI for Data Analytics, prompt engineering and Databricks Machine Learning are listed as preferred skills, not core requirements. Suggested revision exercise: explain a data-analytics use case, how you would assess a prompt's output, and where Databricks Machine Learning might fit; do not assume a specific employer workflow.

Skills to revise

Build a FastAPI endpoint
Suggested practice: write a small, modular Python endpoint that invokes a placeholder model operation; use asyncio where appropriate and explain how it handles concurrent requests. This rehearses the posting's explicit Python, FastAPI and asyncio requirements.
Map an application to AWS
Suggested practice: draft a deployment plan assigning storage, execution, data and container workloads to the named AWS services: S3, Lambda, DynamoDB, ECS and EKS. Explain which services your example actually needs; the posting does not prescribe one architecture.
Provision with Terraform
Suggested practice: write or review a small Terraform configuration for one AWS component in your plan, then explain how you would manage configuration changes. Terraform automation is explicitly required.
Investigate performance with CloudWatch
Suggested practice: use an example of increased response time to identify useful CloudWatch signals and describe how you would narrow down the cause. The posting explicitly mentions monitoring and performance optimization with AWS CloudWatch.
Integrate and check model outputs
Suggested practice: connect the sample endpoint to a replaceable model interface and check its inputs and outputs against a domain-specific use case. Model integration is a stated responsibility; a particular model or evaluation method is not specified.
Review and document Python changes
Suggested practice: review the sample endpoint for clarity, concurrency and failure handling, then write a short description of its interface and limitations. Code reviews, documentation and Python best practices are stated responsibilities.

Your preparation checklist

Prepare a Python GenAI project walkthrough
The role calls for scalable, modular Python applications. Prepare a diagram and a brief account of your design choices, including FastAPI, asyncio and parallel processing where you have used them.
Practice a model-integration design exercise
The posting describes integrating AI models into domain-specific applications with data scientists, ML engineers and product teams. Sketch an API-to-model flow and explain the interfaces and collaboration decisions you would propose.
Draft an AWS deployment plan
The posting names S3, Lambda, DynamoDB, ECS, EKS and Terraform. Draw a deployment diagram, justify which services your example needs, and outline what Terraform would provision.
Prepare a production-readiness review
The role emphasizes scalability, reliability, security and CloudWatch-based performance monitoring. List the risks, signals and improvements you would check for the design you prepared.
Select evidence for preferred skills
Generative AI for Data Analytics, prompt engineering and Databricks Machine Learning are listed as preferred. If you have relevant work, prepare one concrete example of your contribution and its outcome for each applicable area.
Bring a code-review and documentation example
The posting calls for code reviews, documentation and Python best practices. Choose an example showing a specific improvement you proposed or made, plus the documentation needed to maintain it.
Write questions about the team’s work
Ask how the Topaz COE unit chooses domain-specific use cases, divides model-integration work across teams, and evaluates production performance and security. These are questions to clarify the role, not stated company processes.

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…