📫 harshilp1234@gmail.com • 🔗 LinkedIn • 💻 GitHub • 🌐 Portfolio

Final-year B.Tech CSE (AI & ML) undergraduate with a strong interest in applied and emerging technologies. Focused on artificial intelligence, machine learning, and practical software development — building real projects, participating in hackathons, and gaining hands-on experience solving real-world problems.

College of Engineering Kottarakkara, Kerala 📘 B.Tech in CSE (Artificial Intelligence & Machine Learning) — CGPA: 7.77 / 10 (Distinction) 📆 2023 – 2027

Frontend Intern 🏢 MuLearn Foundation — Hybrid 📆 Jan 2026 – Jun 2026 · 6 months

• 🌐 Built responsive frontend features for the MuLearn dashboard using Next.js and React, serving 60,000+ users across the developer community.
• 🧩 Developed reusable UI components with Tailwind CSS and Zustand, establishing a scalable state management architecture across the application.
• ⚡ Integrated RESTful APIs via TanStack Query, reducing load times by 30% and improving data caching efficiency.
• Tech Stack:

Contributor — Festive.js 🏢 Inovus Labs · Hacktoberfest 2025 📆 Oct 2025

• Developed configurable themes and optimized core SDK behavior for Festive.js, a lightweight JavaScript library for dynamic seasonal website overlays.
• Merged multiple PRs collaborating with maintainers; recognized as Top Contributor in the MuLearn Community during Hacktoberfest 2025.
• Tech Stack:

🤟 SignStream – Real-Time Browser-Based Sign Language Recognition · GitHub · Demo · Mar 2026

• Built a browser-based ASL recognition app with live caption buffering and optional text-to-speech (Web Speech API) for accessibility.
• Trained a bidirectional GRU (BiGRU) classifier on 16-frame MediaPipe hand-landmark sequences using WLASL (top-100) plus ASL alphabet data (129-class vocabulary); achieved 97.2% ± 0.25% top-1 accuracy (5-fold stratified CV).
• Deployed a fully client-side inference pipeline using MediaPipe and ONNX Runtime Web (no backend inference), enabling real-time predictions and caption display in the browser.
• 📄 Research paper presented at NCMASE-26 (Mar 2026): "SignStream: A Client-Side Real-Time Browser-Based Sign Language Recognition System"
• Tech:

🧠 Zen – AI Mental Health Companion · GitHub · Demo · Jan 2025

• Developed an AI-powered mental health chatbot with emotion detection, RAG-based knowledge retrieval from 21 clinical psychology PDFs, and interactive wellness tools (breathing exercises, grounding prompts).
• Built embedding-based retrieval using FAISS + Sentence Transformers, grounding LLM responses in evidence-based clinical material.
• Orchestrated Meta-Llama-3-8B-Instruct via Hugging Face Inference API; reduced unsupported responses to ~5% using j-hartmann/emotion-english-distilroberta-base.
• Tech:

🌍 Historical Risk Explorer – NASA Space Apps 2025 · GitHub · Oct 2025

• Built an interactive Streamlit data visualization platform for global environmental datasets with real-time risk probability analysis.
• Improved risk assessment exploration speed by 15% through interactive data streams; led team to Global Nominee selection among thousands of worldwide NASA Space Apps submissions.
• Tech:

🤖 HR Agent – AI-Powered Hiring Automation · GitHub · Sept 2025

• Built a full-stack recruiting automation platform where users paste a job description, upload PDF resumes, and receive AI-ranked candidates — then schedule interviews and send confirmation emails, all from a single UI.
• Architected a Flask REST API backend with a CrewAI agent/crew orchestration layer handling resume parsing, scoring via Google Gemini, and structured candidate ranking.
• Integrated Google Calendar API for real-time slot discovery and event creation (with Google Meet), and SMTP/Gmail for automated interview confirmation emails.
• Frontend built with React + Vite + Tailwind CSS, proxying /api calls to Flask; includes a Dev Mode for mocked behavior without external APIs.
• Tech:

• 🌍 NASA Space Apps Challenge 2025 — Global Nominee — selected among the top global submissions across thousands of international teams.
• 📄 Research Paper — "SignStream: Real-Time Browser-Based Sign Language Recognition via On-Device Inference", NCMASE-26, Mar 2026. · Certificate
• 🎃 Hacktoberfest 2025 — Top Contributor (MuLearn Community) — recognized for significant open-source contributions.
• 🟩 Campus Mantri — GeeksforGeeks (Jan 2026 – Present) · Profile — organized a Campus Awareness Drive reaching 200+ students, growing campus participation in competitive programming.

• 🏅 IBM AI Developer Professional Certificate — IBM
• ☁️ AWS Generative AI and AI Agents with Amazon Bedrock — AWS

Languages & Frameworks

ML / DL / GenAI

DevOps & Tools

• 📄 Resume
• 🤗 Hugging Face
• 🟩 GeeksforGeeks Profile
• 🏅 NCMASE-26 Participation Certificate