Arnav Pandey
@Mlmerchant
Arnav Pandey
@Mlmerchant
Naya Raipur, India
π Hi, I'm Arnav Pandey
π B.Tech in Data Science and Artificial Intelligence at IIIT Naya Raipur (CGPA: 9.29)
π» Machine Learning & AI Enthusiast | π§ Full-Stack ML Developer | π§© Competitive Kaggle Participant
π About Me
I'm passionate about building end-to-end AI systems β from data pipelines and model training to web deployment.
My focus lies in multimodal AI, NLP, computer vision, and generative AI.
I enjoy participating in hackathons, ML competitions, and building tools that combine innovation with real-world utility.
π§ Skills
Languages: Python, C++, SQL
Machine Learning: Scikit-learn, XGBoost, LightGBM, CatBoost, AutoGluon, Optuna
Deep Learning & CV: PyTorch, TensorFlow, Swin V2, ResNet, SupCon, OpenCV, MediaPipe, Albumentations
NLP & GenAI: HuggingFace, LangChain, Gemini API, Ollama, Stable Diffusion
Tools & Infra: Flask, FastAPI, Streamlit, RAPIDS, Pandas, NumPy, Matplotlib, GitHub Actions, Kaggle
π‘ Projects
π Vachan Pay β Voice-Based UPI System
- Built a bilingual voice-controlled UPI assistant using Gemini Flash 2.0 (95%+ intent accuracy).
- Enabled contextual chatbot transactions (check balance, send money, view history).
- Deployed and tested across 6+ Indian accents for speech-to-text robustness.
π§© MindSync β Multi-Modal Mental Health Analysis
- Developed a depression scoring system via Gemini LLM-based essay evaluation.
- Built XGBoost models for emotion prediction from student and song datasets (80% accuracy).
- Designed a YouTube analyzer to assess userβs emotional tone across liked videos and comments.
π SentinalAI β Sentiment Analyzer for YouTubers
- Full-stack app analyzing sentiment from 50+ YouTube videos using RoBERTa (F1 = 0.89).
- Visualized 5,000+ comments via emoji heatmaps, word clouds, and sentiment pie charts.
- Integrated Gemini Flash to auto-generate weekly reports, boosting engagement by 18%.
π Achievements
- π₯ 1st Place β Lazarus Mission ML Hackathon, NIT Karnataka
- π₯ 2nd Place β Humidity & AI Tool Usage Prediction Challenges (MachineHack)
- π₯ 3rd Place β Flight Distance Prediction (MachineHack)
- π 4th Place β NIT Kurukshetra ML Competition
- π§ Top 10 Kaggle Leaderboard Finishes:
- 9th / 4000+ β Predicting Introverts vs Extroverts
- 38th / 2500+ β Optimal Fertilizers Prediction
- π 1 Gold + 7 Bronze Kaggle Medals
π¬ Community & Leadership
- π― Coordinator, Technovateβ25 (Sports): Managed 10-member team; raised βΉ80K+ through registrations.
- π Club Head, Atharva Sports, IIIT NR: Organized 15+ inter-batch tournaments.
- π¨βπ» Kaggle Code Expert: Shared open-source notebooks & guides for ML learners.
π Connect With Me
π§ Email: [email protected]
πΌ LinkedIn: linkedin.com/in/arnav-pandey-96334728a
π GitHub: github.com/letusnotc
π Kaggle: kaggle.com/arnavsuiiiiiiiiiiii
β βTurning data into intelligence, and intelligence into impact.β