Shivansh tiwari

I've built TraceRoot-AI, an agentic incident root-cause analyzer that investigates production failures through logs, metrics, deployment history, and dependency health. Its provider-agnostic pipeline uses Claude and GPT to generate prioritized P0/P1/P2 RCA reports.

At Flipkart GRiD 6.0, I analyzed a 77K+ record travel dataset and engineered demand-forecasting features that improved accuracy from 89% to 91%. I work across Python, machine learning, data analytics, and production-focused system design.

I've also built RateForge, a Java and Spring Boot rate limiter using Redis and atomic Lua scripts to prevent race conditions under concurrent load. Through Nexus, I deployed a full-stack academic resource platform serving 800+ active users and providing access to 300+ research papers.

I'm an active competitive programmer, with 1,000+ DSA and competitive programming problems solved, and I mentor 20+ students through PyC. I enjoy building dependable systems, exploring GenAI workflows, and helping others strengthen their technical foundations.