Risk Analytics and Modeling, Ai Engineering, Full Stack

Full Stack Engineer in Risk Technology working with Modeling and Curve Management using Risk Metrics, Reactjs, TypeScript, C#, SQL Server, Active Batch, Cucumber etc

Leading architecture of scalable LLM systems for interactive storytelling and learning.

• Built RAG pipelines using vector databases for context-aware generation

• Designed prompt-chained, multi-step LLM workflows (story, image, quiz generation)

• Implemented low-latency inference with caching, streaming, and cost optimization

• Developed guardrails (moderation, structured outputs, filtering) for safe AI use

• Architected multimodal pipelines (text, audio, image) at scale

Tech: LLMs, RAG, Vector DBs, Python, TypeScript, Node.js, React, Distributed Systems

Worked as a full-stack developer in the Azure Active Directory team.

● We are the core front end team for Azure Active Directory serving millions of customers and enormous traffic.

● As a senior developer in the team, day to day work involves active coding in various technologies involving web front end (typescript, react etc.) and backend using C#.

● Given the nature of the team, I worked closely with a lot of other teams within the organization in order to design and architect solutions across various new features and navigating around active customers.

● Developed a unique automation flow using MS Flow that helps all PM, Leads, and QA teams across to efficiently understand the execution and burn rate of all active work items / bugs / tests cases and automatically sends periodic communications via email and MS Teams notifications to teams responsible for those tasks. This helped improve the tracking and deliveries of such items over the last few quarters and gives high visibility to higher management.

● Took ownership of various parts within AAD portal and driving their improvements and work towards increased customer

engagement and satisfaction

● Drove the end-to-end work to remove a major dependency on the main backend service that was getting decommissioned and helped move the dependencies to MSGraph and other areas. Deployment was also covered by gradual rollout using ‘Azure Experiments’ to have the least amount to impact to live customers in case of any issues.

I work in Azure Identity team.

Our product is responsible for on-boarding applications onto Azure platform and providing authentication and authorization using Azure Active Directory and Microsoft accounts.

Work involves typical agile process with coding, writing unit/e2e tests, monitoring, deployment to staging and production environment, working closely with other teams to consumes their api and services.

Identity is a challenging space with extremely high learning curve for the product domain and this is something i enjoy the most about this space.