Selected Work

The client required multiple business and customer-facing capabilities to operate within a unified web platform. Instead of managing separate systems for content publishing, events, customer registration, payments, and support, the application needed to bring these workflows together into a single product.

Built as a full-stack Next.js application with server-side logic and Cloudflare D1 for database workflows. The AI support assistant uses the Vercel AI SDK with controlled tool calling (Grok / Cloudflare AI) to invoke predefined application functions for customer inquiries, booking workflows, and delivery-related information.

Application monitoring systems can generate large amounts of logs and event data. Investigating those events manually requires developers to search through stored records to understand what happened, when it happened, which parts of the application were involved, and what patterns exist in the data.

Built with C# and ASP.NET Core for structured API development and high-throughput telemetry ingestion, coupled with MongoDB for persistent monitoring event storage. Semantic Kernel coordinates AI workflows and connects natural-language queries with stored telemetry context.

Language models do not automatically have access to private documents uploaded by users, and large documents can be difficult to search efficiently using traditional keyword matching.

FastAPI backend integrated with LangChain and ChromaDB. Uploaded documents are parsed, chunked, and converted into embeddings stored in ChromaDB. Queries trigger semantic search and an iterative retrieval workflow looping up to 5 times if needed before producing a grounded response.

The project focused on building a functional e-commerce platform that supports core customer and product workflows including catalog search, cart operations, and authentication.

ASP.NET Core RESTful API backend paired with a React frontend, backed by PostgreSQL and Entity Framework Core for transactional reliability.

High-volume applications require reliable, asynchronous notification dispatch without blocking core transaction threads.

C#/.NET backend utilizing message queues and background worker services for asynchronous email and webhook delivery.

Recruiters and job applicants need structured parsing and objective skill matching between resumes and job specifications.

Next.js / Node.js application integrating LLM APIs for structured JSON extraction and skill relevance scoring.