Pranav Vyas

Architected an MCP system via LLM tool-calling over stdio transport, implementing an agentic loop across N concurrent servers and async session lifecycle handling, reducing connection-teardown errors to zero across all client-server pairings. Pioneered a modular Angular frontend leveraging 20+ reusable components and centralized service-layer abstractions for API integration and data interface con

Developed 4-5 TypeScript-based Power BI custom visual components featuring interactive chart rendering pipelines, dataView parsing, categorical and measure mappings, and formatted model bindings, improving analytical workflows. Spearheaded the Power BI custom visual component SDK lifecycle, optimizing visual state management and data change update cycles, viewport recalculations that improved char

Engineered a Recurrent Neural Network (RNN) for temporal forecasting on financial time-series data with exogenous variable integration via rolling window protocol, achieving 80% reduction in Mean Absolute Error (MAE) for closing price predictions.

Grade: 8.88/10 CGPA

Pursued a Bachelor of Technology in Computer Science with a minor in Internet of Things, achieving a CGPA of 8.88/10.