Skip to content
View Aadarshttech's full-sized avatar
🏠
Working from home
🏠
Working from home
6 followers · 6 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull SharkAchievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull SharkAchievement: QuickdrawAchievement: YOLO

Highlights

  • Pro

Organizations

@SahayetaHomeHealth @Kurly-brains @DOAIxILPRL-KathmanduUniversity

Block or report Aadarshttech

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
Aadarshttech/README.md

Hi, I'm Aadarsh Pandit 👋

AI & Full-Stack Developer. I build end-to-end ML pipelines and production-grade web apps. BTech AI, Kathmandu University.

  • 🚀 AI/ML: NLP & ASR, low-resource language fine-tuning, reward-guided training, LLM integration, data curation
  • 💻 Full-Stack: Next.js, React, React Native from backend to polished frontend
  • 🔬 Currently researching: Nepali-English code-switching speech recognition
  • 📄 Lead author of a Whisper fine-tuning paper for low-resource Nepali ASR, accepted (oral) at CHiPSAL 2026
  • 🌱 Always building, always shipping
  • 🤝 Open to: contract work, freelance, research collaborations

🌐 Socials

Instagram LinkedIn Twitter Email Portfolio


🛠️ Tech Stack

🤖 AI / Machine Learning & Data

Python PyTorch TensorFlow scikit-learn Pandas NumPy HuggingFace NLP Kaggle


💻 Web Development (Frontend & Backend)

JavaScript TypeScript React Next JS TailwindCSS NodeJS FastAPI HTML5


🔧 CMS, Tools & Platforms

WordPress Git Docker Linux Flutter


📚 Publications & Research

Reward-Guided Fine-Tuning of Whisper for Low-Resource Nepali ASR
Published at CHiPSAL 2026 Workshop · LREC 2026 · Mallorca, Spain
Developed a reward-guided fine-tuning framework for Whisper utilizing a Random Forest model to filter noisy samples from a 70+ hour dataset, achieving an 11–12% relative improvement in transcription accuracy.

Pinned Loading

  1. Smart-NID-Nepal Smart-NID-Nepal Public

    Making National ID registration effortless. AI scans your citizenship card and instantly fills out the complex bilingual forms so you don't have to.

    TypeScript 3

  2. ai-research-paper-humanizer ai-research-paper-humanizer Public

    The ultimate system prompt for AI assistants (Cursor, Claude, Antigravity) to write and review research papers like a human.

    Python 4

  3. nepali-english-codeswitched-whisper nepali-english-codeswitched-whisper Public

    Data preprocessing and fine-tuning scripts for adapting Whisper to Nepali-English code-switched speech recognition.

    Jupyter Notebook 1

  4. nepali-asr-reward-guided nepali-asr-reward-guided Public

    Official repository for 'Reward-Guided Fine-Tuning of Whisper for Low-Resource Nepali Speech Recognition' (CHiPSAL 2026).

    Python 3 2

  5. nepal-travels-showcase nepal-travels-showcase Public

    An immersive, cinematic web experience showcasing the landscapes of the Himalayas through advanced creative engineering.

    TypeScript 2