Prithviraj R

About

Agentic AI and AI automation developer focused on AI agents, LLM applications, and multi-step workflow orchestration. Builds automation systems with Python, n8n, webhooks, REST APIs, and MongoDB Atlas, and has experience in computer vision and machine learning research.

Experience

• Machine Learning Intern
  Code: https://github.com/prithvirajr203-pixel/Notebook-FlyRank
  Jul 2026 – Aug 2026

  Built a content-refresh prioritization pipeline to flag web pages at risk of going dark from 153,559 performance and 519,606 content records; designed reason codes that turn model outputs into refresh recommendations. Trained and evaluated Logistic Regression and HistGradientBoosting classifiers, reaching 0.87 ROC-AUC on held-out test data.

• AI Research Intern
  National Institute of Technical Teachers Training & Research
  Feb 2026 – Mar 2026

  Engineered a real-time underwater fish-tracking pipeline using YOLOv8, Kalman filtering, and ByteTrack-inspired multi-object tracking to maintain individual fish identity across video frames. Implemented camera-motion compensation and occlusion-handling logic to preserve tracking continuity in dynamic, low-visibility underwater conditions. Developed trajectory mapping and population analytics to support marine research and aquaculture monitoring. Validated model performance against ground-truth footage with research mentors, improving tracking robustness under variable visibility.

Education

• Integrated M.Tech
  Erode Sengunthar Engineering College
  Computer Science and Engineering · 2027
• High School
  Sankar Ponnar Higher Secondary School
  2023
• Secondary
  Sankar Ponnar Higher Secondary School
  2021

Skills

• Computer Vision <3 months
• Python <3 months
• Machine Learning <3 months
• Prompt Engineering <3 months
• JavaScript <3 months
• Model Evaluation <3 months
• OpenCV <3 months
• JSON <3 months
• API Integration Understanding <3 months
• Workflow Automation <3 months
• Webhooks <3 months
• LLM Applications <3 months
• Object Detection <3 months
• AI Agents <3 months
• AI Orchestration <3 months
• Agentic AI Systems <3 months
• APIs and integrations <3 months
• Mediapipe <3 months
• Gesture Recognition <3 months
• Generative AI (ChatGPT, Copilot, Gemini) <3 months
• Kalman Filter <3 months

Tools / apps / platforms

• Git <3 months
• GitHub <3 months
• Google Gmail <3 months
• Groq <3 months
• MongoDB <3 months
• MongoDB Atlas <3 months
• n8n <3 months
• PyCharm <3 months
• Railway <3 months
• Telegram <3 months
• Vercel <3 months
• VS Code <3 months

Languages

• English Conversational Speak

Projects

• AI-Powered IT Support Ticket Automation
  n8n, Groq LLM, Webhooks, REST APIs, Gmail, MongoDB Atlas, Railway

  Architected a 6-workflow n8n automation system covering webhook-based ticket intake, Groq-powered AI classification, urgency detection, and automated routing across support tiers. Built AI agent workflows that interpret incoming support requests, classify issue type and urgency, route each ticket to the correct queue, and generate AI-suggested resolutions for support staff. Integrated Gmail and REST APIs to send automated ticket-created and ticket-resolved emails with MongoDB Atlas as the real-time ticket-tracking backend. Delivered a live dashboard with 30-second auto-refresh for SLA monitoring and deployed on Railway.

• Social Media AI Automation Agent
  Telegram, social platforms

  Built an AI-assisted automation agent that runs daily social-media workflows across Telegram and social platforms, combining scheduled triggers with automated content actions.

• Universal Website Builder Agent
  n8n, AI agents, Vercel

  Engineered an AI-powered website generation system that converts natural-language requirements into structured intake data, AI-generated specifications, UI plans, content, and website code.

• ScamShield AI
  Webhooks, REST APIs

  Designed an AI-powered scam-analysis workflow that applies structured AI reasoning to analyze suspicious messages and generate automated response guidance.

• Gesture YOLO — Free-Hand Alexa for Individuals With Disabilities
  Python, YOLOv8, OpenCV, MediaPipe Hands, pyautogui, speech recognition, pyttsx3

  Designed and implemented a real-time hand-gesture recognition system to enable hands-free computer control for individuals with disabilities and demonstrated real-time control of Gmail, YouTube, and web browsing in a live demo.

Courses & certifications

• Claude 101 · Anthropic
• Introduction to Claude Cowork · Anthropic
• AI Fluency for Builders · CodePath × Anthropic

🏆 Achievements & awards

• 1st Prize, TECHAURORA 2.0 (IEEE KPRIET)

  Won for an AI-powered gesture-recognition system using YOLO.

• Best Paper Award

  Awarded for a research paper presentation at a technical conference.

📚 Publications

• The Agentic Shift: From AI That Talks to AI That Acts · 2026