New opportunity

AI Engineer Digital Solutions

WSP · Bengaluru, India

About this role

WSP is seeking an AI Engineer to develop and support production-oriented Generative AI solutions for the Middle East AI portfolio from the Bengaluru GCC. The role focuses on building AI products, agents, and automation workflows with an emphasis on secure engineering, quality, and responsible AI. Responsibilities: Develop GenAI applications, agents, and automation workflows; implement RAG, function calling, and prompt patterns; integrate AI components with enterprise systems and vector search services; build data pipelines for ingestion, chunking, and embedding; create evaluation datasets for accuracy, safety, and latency testing; apply responsible AI controls including guardrails and human oversight; write clean, documented Python code; and support deployment and monitoring. Qualifications: Bachelor’s degree in Computer Science, Engineering, Data Science or related field; 3+ years of relevant software, data, or AI engineering experience; working knowledge of Python, REST APIs, and SQL; practical understanding of LLMs, embeddings, RAG, and evaluation methods; exposure to Azure AI services, Git, CI/CD, and cloud security. Preferred: Microsoft Azure or AI certifications; experience with agentic AI orchestration, document intelligence, or MLOps. The role operates within a hybrid work model and requires collaboration with global stakeholders and adherence to WSP architecture and governance standards.

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

Skills for this role

PythonGenerative AIRAGPrompt EngineeringAzure OpenAIAzure AI SearchAzure AI Document IntelligenceAzure Machine LearningREST APIsSQLVector SearchGitCI/CDAzure DevOpsMicroservicesCloud Security

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

Production-Grade GenAI Development
The role requires moving beyond prototypes to deliver stable, production-oriented AI solutions. You must demonstrate proficiency in building end-to-end pipelines, including data ingestion, chunking, embedding, and retrieval, while ensuring code is clean, documented, and maintainable within an enterprise environment.
Responsible AI and Governance
Adherence to WSP's strict architecture and governance standards is mandatory. You must prioritize the implementation of guardrails, content filtering, traceability, and human oversight to ensure that all AI agents and workflows operate securely and ethically within the Middle East portfolio.
System Integration and Scalability
You will be responsible for integrating AI components with existing enterprise systems and APIs. This requires a strong grasp of microservices, event-driven patterns, and secure API-first integration to ensure AI solutions function reliably within the broader Azure-based technology stack.
Rigorous Evaluation and Quality Assurance
A core expectation is the creation of evaluation datasets to measure accuracy, latency, and safety. You must be prepared to execute repeatable testing cycles to optimize model performance and ensure that solutions meet the high quality standards required for professional consulting services.

Likely challenges to think through

Balancing Innovation with Strict Governance
WSP operates under specific architectural and security standards. The challenge lies in innovating with new AI patterns while remaining strictly compliant with corporate governance. Prepare by discussing how you have navigated technical constraints in previous projects without sacrificing the quality of the AI output.
Managing Distributed Stakeholder Expectations
Working from a GCC (Global Capability Center) for a Middle East portfolio involves collaborating across different time zones and business cultures. You may face challenges in aligning technical deliverables with the needs of non-technical stakeholders. Prepare by highlighting your experience in transparent reporting and clear communication of technical decisions.
Maintaining Reliability in RAG Pipelines
Retrieval-Augmented Generation (RAG) systems often suffer from latency and grounding issues. The challenge is to maintain high accuracy while scaling. Prepare by discussing your approach to iterative optimization, monitoring, and troubleshooting complex retrieval pipelines in a production environment.

Practice your answers

Technical practice

RAG Pipeline Architecture

What to coverExplain the end-to-end process of building a RAG pipeline, specifically focusing on ingestion, chunking strategies, and embedding generation. Discuss how you would choose an appropriate chunk size and overlap to balance context retention with retrieval accuracy.

Evaluating GenAI Solutions

What to coverHow do you design an evaluation framework for a production-grade LLM application? Address how you measure accuracy, latency, and safety, and describe the role of ground-truth datasets in your testing methodology.

Responsible AI and Guardrails

What to coverWhat specific techniques would you implement to ensure an AI application adheres to responsible AI standards? Discuss the use of content filtering, input/output guardrails, and human-in-the-loop oversight mechanisms.

Agentic AI Orchestration

What to coverDescribe your experience or approach to building agentic workflows where an LLM uses function calling to interact with external enterprise systems. How do you handle error states and ensure the model selects the correct tool for a given task?

Vector Search Optimization

What to coverWhen integrating with services like Azure AI Search, what factors influence the performance and relevance of vector search results? Explain how you would optimize indexing and retrieval for large-scale enterprise data.

Python for AI Engineering

What to coverThe role emphasizes clean, documented, and reusable Python code. How do you structure your AI components to ensure they are maintainable and testable within a CI/CD pipeline?

Managing LLM Latency

What to coverIn a production environment, what strategies would you employ to minimize latency in GenAI applications? Consider aspects like model selection, caching strategies, and asynchronous processing.

Azure AI Integration

What to coverGiven the WSP technology stack, how would you leverage Azure AI services (such as Azure OpenAI or Document Intelligence) to solve a specific business problem? Focus on the integration of these services with existing enterprise data sources.

Behavioural practice

Collaborating with Global Stakeholders

What to coverDescribe a situation where you had to communicate complex technical AI concepts to non-technical stakeholders. How did you ensure alignment on project goals and manage expectations regarding the limitations of the technology?

Handling Production Incidents

What to coverTell me about a time you had to troubleshoot a failure in a deployed AI or data pipeline. What was your process for identifying the root cause, and how did you implement a fix while minimizing downtime?

Adhering to Governance Standards

What to coverProvide an example of a project where you had to strictly follow organizational architecture or security standards. How did you balance the need for innovation with the requirement to comply with established governance?

Iterative Optimization

What to coverDescribe a project where you had to iterate on an AI solution based on performance monitoring or user feedback. What metrics did you track, and how did those insights drive your subsequent development decisions?

Sharpen your knowledge

Topics to prepare

Retrieval-Augmented Generation (RAG) Architecture
The role requires implementing RAG pipelines including ingestion, chunking, and embedding. Exercise: Design a document processing pipeline that handles PDF ingestion, semantic chunking, and vector storage, specifically focusing on how to maintain metadata for traceability and accuracy testing.
Responsible AI and Governance
WSP emphasizes secure engineering and responsible AI controls. Exercise: Research and document a strategy for implementing guardrails, content filtering, and human-in-the-loop oversight for an LLM-based application to ensure compliance with enterprise safety standards.

Skills to revise

Azure AI Services Integration
The role relies on Azure OpenAI, Azure AI Search, and Azure AI Document Intelligence. Exercise: Create a Python script that integrates Azure AI Search with an OpenAI model to perform a semantic search query, demonstrating your ability to handle API authentication and structured output parsing.
Evaluation and Testing for GenAI
You must create evaluation datasets for accuracy, safety, and latency. Exercise: Develop a small test suite using a framework like Pytest to measure the response quality of a prompt against a set of ground-truth answers, focusing on metrics like hallucination rate and response time.
CI/CD for AI Workflows
The position requires deployment and monitoring within an Azure DevOps environment. Exercise: Draft a basic CI/CD pipeline configuration (YAML) that automates the testing of Python-based AI components whenever code is pushed to a Git repository.

Your preparation checklist

Develop a RAG pipeline project
Build a Python-based application that ingests documents, chunks them, and stores them in a vector database. Demonstrate your ability to implement retrieval-augmented generation using Azure AI services or similar frameworks, ensuring you can explain your choice of embedding models and retrieval strategies.
Prepare an evaluation framework walkthrough
Create a sample evaluation dataset and a testing script that measures accuracy, latency, and safety guardrails for an LLM application. Be ready to discuss how you would implement human-in-the-loop oversight and content filtering to meet enterprise responsible AI standards.
Review Azure AI and MLOps best practices
Familiarize yourself with the WSP-aligned stack, specifically Azure OpenAI, Azure AI Search, and Azure DevOps for CI/CD. Prepare to discuss how you manage the model lifecycle, from prototype to production monitoring, using automated testing and observability tools.
Formulate questions for the interviewers
Prepare thoughtful questions regarding the team's current approach to agentic AI orchestration, how they balance global stakeholder requirements with local GCC delivery, and the specific governance standards applied to their production-oriented AI solutions.

Availability has not yet been verified. 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…