Skip to content
View monishmal3375's full-sized avatar

Block or report monishmal3375

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
monishmal3375/README.md

Hey, I’m Monish

ML Engineer | Full-Stack AI Developer | Open-Source Builder

I build production-grade AI systems. Focused on voice AI, LLM security, and end-to-end AI applications.


🚀 Featured Projects

NOVA-VAD 🎙️

Production-Grade Voice Activity Detection

  • 93% accuracy on real-world noisy audio
  • Real-time streaming inference (<50ms latency)
  • Noise-robust (background chatter, wind, AC hum)
  • 3-class detection: Speech, Noise, Silence
  • Fully explainable model architecture
  • Benchmarked on 5+ datasets
  • Open-source, MIT licensed

→ View Repository | → Star ⭐


Vexom 📱

AI-Powered iOS Academic Assistant

An intelligent study companion built with SwiftUI and advanced AI architecture.

Tech Stack: SwiftUI | Claude API | Firebase | ChromaDB | RAG

Features:

  • Socratic Tutor - AI-guided learning with adaptive questioning
  • Exam Predictor - Analyze patterns and predict performance
  • SemesterArc - GPA tracking and semester analytics
  • DeepWorkView - Deep work session tracking and insights
  • WeeklyBriefing Engine - Automated weekly summaries

Production-ready iOS app with real-time AI interactions and secure authentication.

→ View Repository


NomadAI ✈️

Full-Stack AI Travel Planner

End-to-end travel planning application powered by Claude AI and built with modern web technologies.

Tech Stack: Next.js | TypeScript | NextAuth | Claude API | Vercel

Features:

  • AI-powered itinerary generation
  • Real-time travel recommendations
  • Personalized trip planning
  • Secure authentication with NextAuth
  • Responsive design across devices

Demonstrates full-stack proficiency from backend API integration to production-grade frontend.

→ View Repository


💼 Current Focus

Area Details
Voice AI Building production VAD systems, audio processing, real-time inference
LLM Security Securing LLMs, RAG system architecture, prompt injection prevention
Full-Stack AI End-to-end AI applications with modern frameworks (Next.js, SwiftUI)
Open-Source Contributing to and building maintainable, well-documented projects

🛠 Tech Stack

Languages: Python, TypeScript, Swift, Java, SQL

AI/ML: PyTorch, TensorFlow, Hugging Face, Claude API, RAG Systems

Frontend: React, Next.js, SwiftUI, Tailwind CSS

Backend: Python, Node.js, Firebase, REST APIs

Data & DevOps: Azure Databricks, SQL, Git, GitHub

Tools: VSCode, Xcode, Jupyter, Docker


🎓 Background

  • Cybersecurity Sophomore @ Purdue University
  • ML & Full-Stack AI focused developer
  • Published Research on AI Cancer Diagnosis
  • Production Experience at AI startups and tech companies

🔗 Let’s Connect

  • 🐦 Twitter/X - Shipping projects in public, voice AI deep dives
  • 💼 LinkedIn - Professional background

🎯 Open To

  • Collaborations on open-source ML projects
  • Discussions on LLM security and RAG systems
  • Speaking about building AI products
  • Connecting with builders and engineers

📈 Project Stats

Total Repositories: 15+
Languages: 6+ (Python, TypeScript, Swift, Java, SQL, More)
Open-Source Projects: Multiple
Active Development: Ongoing

🌟 Recent Work

Recently shipped:

  • NOVA-VAD - Production voice activity detection (93% accuracy)
  • Multiple full-stack AI applications
  • Contributions to open-source ML ecosystem

Currently building:

  • Advanced voice AI systems
  • LLM security research
  • Full-stack AI products

💡 Philosophy

I believe in:

  • Building in public - shipping early and often
  • Production-first - code that works in the real world
  • Open-source - contributing back to the community
  • Technical depth - understanding systems deeply
  • End-to-end ownership - full-stack responsibility

Last updated: June 2026 | Always building 🚀

Pinned Loading

  1. nova-vad nova-vad Public

    Lightweight, explainable VAD that beats Silero and WebRTC on noisy audio. No GPU required.

    Python 10 1

  2. Vexom Vexom Public

    Swift

  3. NomadAI NomadAI Public

    TypeScript