New opportunity

AI Engineer_Full Stack Generative AI

Burns & McDonnell · Bengaluru, India

About this role

We are looking for an AI Engineer with Full-Stack experience to build scalable, production-ready Generative AI and Agentic AI applications across frontend, backend, AI, and cloud. The ideal candidate will independently manage the full lifecycle of a solution from UI development to backend integration, AI agent design, cloud deployment, and production monitoring. Key Responsibilities: Develop responsive frontend applications using React, Angular, or Svelte. Build scalable backend services and REST APIs using Python and FastAPI. Design and develop AI agents and multi-agent workflows using Google ADK, LangChain, and LangGraph. Build RAG solutions using embeddings, vector search, metadata filtering, and reranking. Work with Vector Databases such as Vertex AI Vector Search, pgvector, Pinecone, Qdrant, or Weaviate. Integrate Gemini, Vertex AI, and enterprise APIs into applications. Build document ingestion and knowledge-processing pipelines. Deploy scalable applications on GCP, including Cloud Run, Pub/Sub, GCS, Cloud SQL, and Secret Manager. Implement secure authentication using OAuth 2.0, JWT, OIDC, and IAM. Build asynchronous/event-driven services and optimize applications for performance, scalability, reliability, and cost. Implement testing, monitoring, logging, CI/CD, and production support. Required Qualifications: B.Tech/B.E, BCA, or BSC in any specialization. 5 to 8 years of experience. Strong software engineering, cloud, and Generative AI fundamentals. Good to Have: Experience with MCP (Model Context Protocol), Microsoft Graph API, Azure Entra ID, AI evaluation and observability, OpenTelemetry, Terraform, WebSockets, SSE, and enterprise AI applications.

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

Skills for this role

ReactAngularSvelteJavaScriptTypeScriptPythonFastAPIREST APIsPostgreSQLGeminiVertex AIGoogle ADKLangChainLangGraphRAGVector DatabasesPrompt EngineeringGCPCloud RunPub/SubGCSCloud SQLDockerCI/CDIAM

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 Lifecycle Ownership
The role requires the ability to independently manage the full development lifecycle, from UI design and backend API development to AI agent orchestration and cloud deployment. Candidates must demonstrate proficiency in bridging the gap between frontend interfaces and complex backend AI workflows.
Agentic AI and RAG Implementation
A core priority is designing and deploying multi-agent workflows using LangChain, LangGraph, and Google ADK. This includes building robust RAG pipelines that incorporate advanced techniques like metadata filtering, reranking, and vector database management to ensure accurate knowledge retrieval.
Scalable GCP Cloud Architecture
Engineers must be capable of deploying production-ready applications on Google Cloud Platform. This involves utilizing services like Cloud Run, Pub/Sub, and Cloud SQL, while ensuring security through IAM, Secret Manager, and OAuth 2.0 protocols.

Likely challenges to think through

Balancing Full-Stack Breadth with AI Depth
The role demands high proficiency in both traditional web development (React/Angular/Svelte) and specialized AI engineering. The challenge lies in maintaining high-quality frontend performance while simultaneously optimizing complex, asynchronous AI agent logic and document ingestion pipelines.
Productionizing Non-Deterministic AI Systems
Moving from prototype to production-ready Generative AI involves significant hurdles in reliability and observability. Prepare by discussing strategies for implementing logging, monitoring, and testing frameworks that account for the non-deterministic nature of LLM outputs and agentic workflows.

Practice your answers

Technical practice

Designing Multi-Agent Workflows with LangGraph

What to coverExplain how you would structure a multi-agent system using LangGraph to handle complex tasks. Discuss the importance of state management, cyclic graphs, and how you ensure individual agents maintain context while interacting with external tools.

Optimizing RAG Pipelines for Accuracy

What to coverDescribe your approach to improving RAG performance beyond basic retrieval. Focus on techniques like metadata filtering, reranking strategies, and how you handle document ingestion to ensure high-quality semantic search results.

Scalable Backend Architecture with FastAPI

What to coverHow do you design a scalable, asynchronous backend using FastAPI to support high-concurrency AI applications? Discuss your strategy for handling long-running AI inference tasks without blocking the main event loop.

GCP Deployment and Infrastructure Management

What to coverDetail your experience deploying containerized applications to Google Cloud Run. Explain how you manage secrets, configure IAM roles for least privilege, and utilize Pub/Sub for event-driven service communication.

Vector Database Selection and Trade-offs

What to coverCompare different vector databases like Pinecone, Qdrant, or pgvector. Suggest criteria for choosing one over the others based on factors like latency, scalability, cost, and existing infrastructure integration.

Implementing Secure Authentication in AI Services

What to coverExplain how you would implement OAuth 2.0 and OIDC for an enterprise AI application. Discuss the security considerations for managing JWTs and ensuring secure access to sensitive enterprise APIs.

Production Monitoring and Observability for AI

What to coverHow do you monitor the health and performance of Generative AI applications in production? Suggest tools or methodologies for tracking latency, token usage, and model output quality (observability).

Frontend Integration with AI Backends

What to coverDescribe how you handle real-time AI responses on the frontend using React or Svelte. Discuss the use of WebSockets or Server-Sent Events (SSE) to stream token generation to the user interface.

Behavioural practice

Managing the Full Lifecycle of an AI Solution

What to coverDescribe a time you independently owned a project from UI design through to production deployment. Focus on how you balanced technical debt with feature delivery and managed the end-to-end lifecycle.

Handling Technical Trade-offs in System Design

What to coverProvide an example where you had to choose between two different architectural approaches for a scalable service. Explain your decision-making process, the trade-offs you considered, and the final outcome.

Collaborating on Complex Technical Challenges

What to coverTell me about a situation where you had to explain a complex AI or cloud architecture concept to non-technical stakeholders. How did you ensure alignment and move the project forward?

Adapting to New AI Frameworks

What to coverDescribe a situation where you had to quickly learn and implement a new AI framework or tool (e.g., LangGraph or a new vector database) to meet a project requirement. How did you validate your implementation?

Sharpen your knowledge

Topics to prepare

Multi-Agent Workflow Orchestration
The role requires designing agentic AI systems using LangGraph. Revision exercise: Design a multi-agent architecture where one agent handles document retrieval and another handles synthesis/reranking, documenting the state transitions and tool-calling logic. This is directly relevant to the requirement for building complex, scalable agentic workflows.
RAG Pipeline Optimization
The role emphasizes RAG solutions including metadata filtering and reranking. Revision exercise: Create a document ingestion pipeline that processes raw text into chunks, generates embeddings, and implements a reranking step to improve retrieval precision. This addresses the core requirement for building production-ready knowledge-processing pipelines.

Skills to revise

FastAPI Backend Development
The role requires building scalable REST APIs using Python and FastAPI. Revision exercise: Build a containerized FastAPI service that integrates with a vector database and implements OAuth 2.0/JWT authentication. This is essential for the full-stack responsibilities of the role.
GCP Cloud Deployment
The role requires deploying applications on GCP using Cloud Run and managing secrets. Revision exercise: Create a CI/CD pipeline using GitHub Actions or Google Cloud Build to deploy a containerized application to Cloud Run, utilizing Secret Manager for environment variables. This maps directly to the deployment and DevOps requirements.
Vector Database Integration
The role requires working with databases like Vertex AI Vector Search or pgvector. Revision exercise: Implement a CRUD interface for a vector database that supports semantic search and metadata filtering. This is a critical hands-on skill for the advertised RAG and AI agent responsibilities.

Your preparation checklist

Architect an End-to-End RAG Pipeline
Prepare a technical walkthrough of a project where you built a RAG system using Python, FastAPI, and a vector database (e.g., Pinecone or pgvector). Be ready to explain your choices for embedding models, metadata filtering, and reranking strategies to ensure high-quality retrieval.
Demonstrate Agentic Workflow Expertise
Prepare a case study showcasing a multi-agent system developed with LangChain or LangGraph. Focus on how you implemented tool/function calling and managed state across asynchronous, event-driven workflows to solve complex, multi-step tasks.
Map GCP Deployment and Monitoring Strategy
Outline your approach to deploying production-ready AI applications on GCP. Prepare to discuss how you utilize Cloud Run for scalability, manage secrets via Secret Manager, and implement observability using logging, monitoring, and OpenTelemetry to track AI performance.
Prepare for Full-Stack Integration Discussions
Review your experience connecting frontend frameworks like React or Svelte to FastAPI backends. Be prepared to discuss how you handle secure authentication (OAuth 2.0/OIDC) and maintain responsive UI performance when integrating real-time AI responses via WebSockets or SSE.
Formulate Strategic Questions for the Interviewer
Develop thoughtful questions regarding the team's current AI evaluation frameworks, how they handle document ingestion at scale, and the specific enterprise challenges they face when integrating Gemini and Vertex AI into their existing infrastructure.

Availability last checked: 18 Oct 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…