Krishna Katiyar

About

Computer Science undergraduate building full-stack software across backend systems, machine learning pipelines, and product interfaces. Experienced in AI/ML, backend development, and research projects spanning neuroimaging, policy engines, and predictive control systems.

Experience

• Summer Research Intern
  Centre of NeuroInformatics, VIT Chennai
  May 2026 – Jul 2026

Education

• B.Tech in Computer Science and Engineering, AI and Robotics
  Vellore Institute of Technology, Chennai
  Computer Science and Engineering, AI and Robotics · 2024 – 2028

Skills

• SQL <3 months
• C <3 months
• Computer Vision <3 months
• Deep Learning <3 months
• Express.js <3 months
• FastAPI <3 months
• Python <3 months
• CSS3 <3 months
• Java <3 months
• JavaScript <3 months
• Node.js <3 months
• Statistics <3 months
• NumPy <3 months
• Pandas <3 months
• Streamlit <3 months
• scikit-learn <3 months
• OpenCV <3 months
• React <3 months
• PyTorch <3 months
• TensorFlow <3 months
• TypeScript <3 months
• ESP32 <3 months
• LangChain <3 months
• WebSockets <3 months
• JWT <3 months
• API Integration Understanding <3 months
• HTML5 <3 months
• RAG <3 months
• RBAC <3 months
• ROS <3 months
• SQLAlchemy <3 months
• Sentence Transformers <3 months

Tools / apps / platforms

• Amazon Web Services AWS <3 months
• Arduino <3 months
• Docker <3 months
• Git <3 months
• Hugging Face Transformers <3 months
• Linux <3 months
• Microsoft Power BI <3 months
• Netlify <3 months
• PostgreSQL <3 months
• SQLite <3 months
• Tableau <3 months
• Vercel <3 months

Projects

• Alzheimer Trajectory Net
  MRI, Deep Learning

  Built a longitudinal MRI trajectory prediction pipeline processing 3D imaging tensors with preprocessing, augmentation, and class-imbalance handling. Evaluated using macro F1-score and confusion-matrix analysis.

• ThermoPilot - Predictive HVAC Optimization
  Model Predictive Control, CVXPY, Streamlit

  Designed an MPC system for HVAC scheduling using a 2R2C building thermal physics model and a gradient-boosted forecasting layer. Formulated HVAC energy cost minimization as a convex quadratic program and built a Streamlit dashboard for cost-vs-comfort visualization.

• Guardian - AI Security Risk Assessment & Policy Engine
  Sentence Transformers, FastAPI, PostgreSQL, React, TypeScript

  Built an AI-powered security risk assessment platform using deterministic policy rules and Sentence Transformers to classify sensitive actions into Allow, Review and Block decisions. Implemented explainable risk scoring, JWT/RBAC authentication, rate limiting, analysis ownership, audit logging, and a React and TypeScript control plane.

Courses & certifications

• Make Agentic AI Work for You · IBM
• Retrieval-Augmented Generation for Enhanced AI Outputs · IBM
• AWS Foundations: Getting Started with the AWS Cloud Essentials · Amazon Web Services
• AWS Cloud Practitioner Essentials · Amazon Web Services
• Machine Learning Specialization · Stanford Online & DeepLearning.AI

🏆 Achievements & awards

• Winner, Makeathon 2025 · 2025

  Won among 60+ teams and led AI model development using CodeT5 and CodeBERT.