Skip to content
View tannnmayy's full-sized avatar
🏠
Working from home
🏠
Working from home
10 followers · 48 following

Achievements

Achievement: YOLOAchievement: Pull Shark

Achievements

Achievement: YOLOAchievement: Pull Shark

Highlights

  • Pro

Block or report tannnmayy

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
tannnmayy/README.md

Hi, I'm Tanmay Singh 👋

Software Developer | AI/ML Researcher | Building Lightweight Deep Learning Systems

🚀 About Me

  • 🎓 B.Tech CSE (Big Data Analytics) @ SRM Institute of Science and Technology, KTR — Class of 2028
  • 🔬 Samsung PRISM Research Intern @ SRI-B, building lightweight deep learning ISPs that denoise and HDR tone map at the same time in under <6000 Parameters and 30fps for mobile camera systems
  • 🤝 President & Co-founder, FAST SRM — NVIDIA Student Developer Network club driving applied AI research with Nissan, HAL, DRDO, and CMRL
  • 🏆 1st place, HACK-SUMMIT Hackathon, out of 400+ teams
  • 🏆 1st place, WiCyS SRMIST Hackathon, out of 200+ teams
  • 🏆 3rd place, HYBRID-HACK Hackathon, out of 150+ teams
  • 🏆 Top-10, Schneider Electric Hackathon, out of 100+ teams
  • 📜 2× AWS Certified · Oracle AI Foundations Certified
  • 🌱 Building toward open-source contributions via LFX Mentorship (CNCF)
  • 💼 Open to Software Engineering, AI/ML, and Research internships

🔬 Current Work — TinyISPNet

Designing a 5.8K-parameter neural ISP for the Sony IMX135 camera sensor that jointly performs RAW image denoising and HDR tone mapping, targeting 30+ FPS on mobile GPUs. The pipeline takes raw Bayer measurements straight to perceptually optimized RGB output, trained end-to-end with an L1 + SSIM perceptual objective across 2,120+ RAW/JPEG image pairs.

🛠️ Tech Stack

Languages & Frameworks

AI / ML & Data

Databases

Tools & DevOps

--- ### 📊 GitHub Stats

GitHub Stats

GitHub Streak

📫 Let's Connect

Pinned Loading

  1. AI-Swarm AI-Swarm Public

    you cant run a open weights model on your machine alone so now you connect multiple devices (like your phone and laptop) available in your vicinity all via a browser only application and run any mo…

    JavaScript

  2. aqi-sentinel- aqi-sentinel- Public

    Python

  3. neuroread-proj neuroread-proj Public

    NeuroRead is an AI-powered adaptive reading system that transforms complex digital content into accessible, easy-to-understand formats. Combining a personalized web platform with learning and assis…

    TypeScript

  4. DepGraph DepGraph Public

    A high-performance Python CLI to detect dependency conflicts in monorepos. It scans your project in <1s, simulates upgrade impacts, and generates interactive D3.js visualizations. Built with Networ…

    Python 1

  5. neurolearn-chrome-extension- neurolearn-chrome-extension- Public

    Neuro-Reads is a Chrome extension that makes web pages easier to read for people with dyslexia. It applies an accessible font, larger text, and soft backgrounds, then uses an AI model via Hugging F…

    JavaScript