Ananya Katiyar

I’m an AI / Machine Learning Engineering Intern and B.Tech (Computer Science Engineering) student focused on turning models into production-ready systems. I enjoy engineering high-performance pipelines where latency, accuracy, and reliability all matter.

At IIT Mandi, I engineered high-performance ML pipelines in C++ for real-time image recognition, processing 100K+ samples with sub-10ms latency. I improved prediction accuracy by 18% through feature engineering and hyperparameter tuning, and reduced end-to-end inference time by 35% using multi-threading and cache-efficient memory management.

I also built scalable graph processing systems at NIT Rourkela, managing datasets exceeding 1M+ nodes. By implementing BFS, DFS, and an A* heuristic, I accelerated runtime by 28%, and supported faculty review by authoring 3 technical reports after profiling across 10+ datasets.

In my projects, I ship practical systems: SecureCode Analyzer integrates LLMs with AST parsing to detect 15+ OWASP vulnerability categories and uses CI/CD incremental diff scanning to cut redundant scans by 60%. I’ve also worked on LoadGuard’s Redis token-bucket rate limiting (5,000+ requests/min with sub-20ms p99 latency) and an AI Virtual Health Assistant achieving a 92% resolution rate on 500+ simulated queries.