New opportunity

AI/ML Engineer

Wissen Technology Pvt. Ltd. · Bangalore South, India

About this role

Wissen Technology is seeking an AI/ML Engineer to build generative AI applications and agentic AI solutions. This full-time, hybrid role is based in Bangalore; the posting lists Bangalore South, India. The stated experience range is 2–5 years, and the education qualification is MCA, B.E., B.Tech or M.Tech.

Responsibilities

Design and develop generative AI applications using LLMs and modern AI frameworks. Build agentic solutions involving autonomous agents, multi-agent workflows, tool calling, reasoning and task orchestration.
Integrate AI capabilities into enterprise applications and business workflows, working with data, APIs, knowledge sources and enterprise systems.
Design scalable, production-ready AI architectures with attention to performance, reliability and security. Collaborate with product, engineering and business teams to turn use cases into solutions.
Evaluate AI models and improve their accuracy, relevance and performance.

Preferred skills and experience:

Hands-on experience with generative AI applications and proven experience building agentic solutions; understanding of LLMs, prompt engineering, AI application architecture and integrations; and experience developing and deploying production-grade AI solutions.
Experience with Python and AI/ML frameworks; exposure to LLM platforms, AI APIs, cloud AI services or open-source AI models; understanding of APIs, microservices, databases and cloud platforms; and strong problem-solving and analytical skills.

Additional qualifications: NLP, machine learning, ontology and knowledge representation, knowledge graphs, RAG, vector databases, embeddings and AI-powered search are listed as good to have. Computer vision, IoT or connected systems, and forecasting or predictive analytics are listed as nice to have.

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

Skills for this role

Generative AIAgentic AILarge Language ModelsPrompt EngineeringAI Application ArchitectureAI IntegrationAutonomous AgentsMulti-Agent WorkflowsTool CallingTask OrchestrationAI Model EvaluationPythonNatural Language ProcessingMachine LearningOntologyKnowledge RepresentationKnowledge GraphsRAGVector DatabasesEmbeddingsAI-Powered SearchComputer VisionIoTForecastingPredictive AnalyticsAPIsMicroservicesDatabasesCloud PlatformsCloud AI Services

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 generative AI applications
The role calls for designing and developing LLM-based applications with modern AI frameworks. Prepare to explain how you would structure an application, from model interaction and prompts through deployment.
Engineer agentic workflows
Autonomous agents, multi-agent workflows, tool calling, reasoning and task orchestration are explicit duties. Prepare to explain when an agentic design is useful and how its steps and tool use would be controlled.
Integrate AI into enterprise workflows
The posting emphasizes connecting AI capabilities with business workflows, data, APIs, knowledge sources and enterprise systems. Prepare an integration example that makes the data flow and system boundaries clear.
Deliver and improve production-ready solutions
The role combines scalable architecture, performance, reliability and security with evaluating models for accuracy and relevance. Prepare to describe how you would assess output quality and use findings to improve a deployed application.

Likely challenges to think through

Balancing agent autonomy and control
Preparation hypothesis: more autonomous, multi-step workflows may handle complex tasks but make tool use and failures harder to manage. Be ready to explain how you would limit actions, observe execution and handle unsuccessful steps.
Trading output quality against production constraints
Preparation hypothesis: improvements in accuracy or relevance may affect latency, cost or architectural complexity, while the posting requires performance and reliability. Prepare a concrete way to compare options against use-case needs rather than optimizing one metric alone.
Turning business use cases into workable integrations
Preparation hypothesis: product, engineering and business teams may frame needs differently, and enterprise data and APIs can constrain the solution. Prepare to explain how you would clarify the intended outcome, identify required systems and agree on how success will be evaluated.

Practice your answers

Technical practice

How would you design a generative AI application around an LLM?

What to coverThe role involves designing and developing LLM-based applications. Suggested preparation: Explain how you would define the task, structure prompts and application logic, and test whether outputs meet the use case.

When would you use one agent rather than a multi-agent workflow?

What to coverThe posting names autonomous agents, multi-agent workflows, reasoning and task orchestration. Suggested preparation: Compare coordination complexity, reliability and the value of splitting a task among agents.

How would you design tool calling for an agent?

What to coverTool calling and security are both part of the stated work. Suggested preparation: Discuss how an agent chooses and invokes tools, handles failures, and limits access to actions or data.

How would you integrate an AI capability into an enterprise workflow?

What to coverThe role involves enterprise applications, data, APIs, knowledge sources and enterprise systems. Suggested preparation: Walk through interfaces, data flow, error handling and how you would fit the capability into an existing workflow.

When would retrieval-augmented generation improve an LLM application?

What to coverRAG, embeddings, vector databases and AI-powered search are listed as good to have. Suggested preparation: Explain retrieval and generation stages, when external knowledge helps, and how you would check whether retrieved material is relevant.

How would you choose an LLM platform or model for an application?

What to coverThe posting mentions exposure to LLM platforms, AI APIs, cloud AI services or open-source models. Suggested preparation: Compare options against the use case's quality, integration, performance and security needs without assuming a particular provider.

What would you consider when taking an AI application into production?

What to coverThe role calls for scalable, production-ready architecture and deployment experience. Suggested preparation: Describe how you would assess performance, reliability and security together, including trade-offs when demand grows.

How would you evaluate whether a model change improved an AI application?

What to coverEvaluating models and improving accuracy, relevance and performance are explicit responsibilities. Suggested preparation: Explain how you would select representative tasks, compare results and check that gains in one measure do not mask regressions in another.

How would you troubleshoot a decline in response relevance?

What to coverThe posting calls for continuous improvement of relevance and strong analytical problem-solving. Suggested preparation: Describe how you would isolate possible causes in inputs, prompts, knowledge sources or the model, then verify a fix against examples.

How would you structure a Python-based AI service?

What to coverPython, AI/ML frameworks, APIs and microservices appear in the preferred experience. Suggested preparation: Outline the boundaries between AI logic and application interfaces, and explain how that structure supports testing and maintenance.

Behavioural practice

Tell me about a time you turned an ambiguous use case into a technical plan.

What to coverThe role involves translating use cases into AI solutions with product, engineering and business teams. Suggested practice: Give the situation, how you clarified the need and chose a direction, and the result.

Describe a time you resolved a disagreement across technical and business stakeholders.

What to coverCross-functional collaboration is an explicit responsibility. Suggested practice: Explain the situation, how you surfaced trade-offs and reached a decision, and what resulted.

Tell me about a difficult problem you diagnosed in an application you built.

What to coverThe posting emphasizes complex AI use cases, analytical problem-solving and reliability. Suggested practice: Describe the situation, your diagnostic actions and the outcome, distinguishing evidence from initial assumptions.

Describe a time evaluation results led you to revise a solution.

What to coverThe role includes evaluating models and improving accuracy, relevance and performance. Suggested practice: State the situation, what the results showed, what you changed and the result.

Tell me about a time you balanced delivery speed with production quality.

What to coverThe posting calls for production-ready AI applications with attention to performance, reliability and security. Suggested practice: Describe the situation, the trade-off you made, how you communicated it and the result.

Sharpen your knowledge

Topics to prepare

LLMs and prompt engineering
Posting: LLM-based applications and prompt engineering are preferred. Suggested revision exercise: explain how you would structure a prompt for a business task, then identify two likely failure modes and how you would assess the output.
Agentic workflow design
Posting: autonomous agents, multi-agent workflows, tool calling, reasoning and task orchestration are responsibilities. Suggested revision exercise: diagram a workflow that decides when to call a tool, passes results between steps and handles a failed call.
Enterprise AI integration
Posting: the role integrates AI with business workflows, data, APIs, knowledge sources and enterprise systems. Suggested revision exercise: sketch the data flow from an application request through an AI capability to an enterprise API, noting where access checks belong.
Production AI architecture
Posting: architectures must address scalability, performance, reliability and security. Suggested revision exercise: compare two designs for handling concurrent AI requests and explain their reliability and security trade-offs.
Model evaluation
Posting: evaluate models and improve accuracy, relevance and performance. Suggested revision exercise: define a small set of representative requests, expected outcomes and criteria for comparing responses before and after a change.
Retrieval and knowledge representation
Posting: RAG, embeddings, vector databases, AI-powered search and knowledge graphs are good to have, not core requirements. Suggested revision exercise: explain when retrieval from a knowledge source could improve an answer and how you would check whether the retrieved material is relevant.

Skills to revise

Build a Python LLM application
Posting: hands-on generative AI development and Python with AI/ML frameworks are preferred. Suggested practice: implement a small Python request-to-response flow using an AI API or model available to you, and inspect how prompt changes affect its output.
Implement tool-calling control flow
Posting: building agentic solutions with tool calling and task orchestration is a responsibility. Suggested practice: write a small flow that selects between two callable functions, validates their inputs and handles a failed call.
Connect AI output to an application API
Posting: the role works with APIs, data and enterprise applications. Suggested practice: connect a sample AI response to a sample API request, checking input validation, response handling and what data is passed onward.
Run a repeatable evaluation
Posting: continuously improving accuracy, relevance and performance is a responsibility. Suggested practice: create a small test set, record outputs for two prompt or model configurations, and summarize which changes improved or worsened the defined criteria.
Exercise deployment and failure handling
Posting: production-grade deployment, reliability, performance and security matter. Suggested practice: walk a small AI application through a deployment checklist, test an unavailable API and excessive request volume, and document how the application responds. This is preparation practice, not a stated employer process.

Your preparation checklist

Prepare a generative AI project walkthrough
The role calls for hands-on LLM application development. Prepare a concise walkthrough of a project you can discuss, covering the use case, your contributions, framework and model choices, prompts, results, and limitations.
Map out an agentic workflow
The posting emphasizes autonomous agents, multi-agent workflows, tool calling, reasoning, and orchestration. Prepare a diagram of an agentic solution you built or a clearly labeled design exercise, showing task flow, tool interfaces, and how failures are handled.
Practice an enterprise integration design
The role involves data, APIs, knowledge sources, and business workflows. Sketch how an AI feature could connect to an enterprise application; explain your proposed architecture and trade-offs for performance, reliability, and security.
Bring evidence of production engineering
The posting prefers Python, AI/ML frameworks, and experience developing and deploying production-grade solutions. Select specific examples you can substantiate and prepare to explain implementation, deployment, and operational decisions without overstating your role.
Prepare an evaluation plan
Model evaluation and improvement are stated responsibilities. Draft a small test set and describe how you would assess accuracy, relevance, and performance, investigate failures, and decide whether a change helped.
Identify relevant optional experience
RAG, embeddings, vector databases, and knowledge graphs are listed as good to have, not core requirements. If applicable, prepare one brief example with its retrieval or representation choices and trade-offs; otherwise, frame it as an area to discuss rather than claimed experience.
Write questions for the employer
Prepare questions about the AI use cases, available data and integrations, how teams evaluate deployed solutions, and how product, engineering, and business stakeholders collaborate. These follow the responsibilities in the posting without assuming a particular company process.

Availability last checked: 5 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…