Skip to content
View AP4549's full-sized avatar
6 followers · 9 following

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: Quickdraw

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: Quickdraw

Highlights

  • Pro

Block or report AP4549

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.

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
AP4549/README.md

👾 Ayush Pandita.exe

🧠 About Me

🎓 4th Year B.Tech Data Science Student @ NMIMS (MPSTME)
💡 Focused on AI, Machine Learning & Financial Technology
🚀 Building multi-agent systems & intelligent automation platforms

  • 🔭 Currently working on: AI Customer Support Systems
  • 🌱 Exploring: Quant Finance + LLM Architectures
  • ⚡ Philosophy: Build systems that decide, not just predict

🔗 Connect with Me

🚀 Projects

🤖 AI Customer Support System (Multi-Agent)

  • Multi-agent system using Ollama (on-prem LLM)
  • Automates summarization, routing, resolution
  • Tech: React + Flask + SQLite

📊 Forex Future Prediction

  • ML-based FX forecasting system
  • Focus on analytics + visualization

🧠 ParkiSense

  • AI-powered Parkinson’s detection system
  • Uses keystroke + voice + chatbot

📈 Chart Whisper

  • AI tool to analyze charts & answer questions
  • Built using TypeScript + AI

⚽ Premier League Classification

  • CNN-based sports image classification
  • Focus on model benchmarking

📄 Research Work

🧾 IEEE Conference Publication

An End-to-End AI Framework for Multilingual Financial Assistance and Inclusion
📍 2025 International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)
🏢 Publisher: IEEE
📌 DOI: https://doi.org/10.1109/ICRCICN68210.2025.11364961

Authors: Ayush Pandita, Nathan Dsouza, Emil Pereira, Kshamta Mathur

🔍 Highlights:

  • Built Credit Mate (multilingual AI financial assistant)
  • Integrated Whisper + Translate API + Fine-tuned LLM
  • 🚀 +30% accuracy improvement (LoRA fine-tuning)
  • ⚡ Low latency: 1.8–3.2s
  • 🌍 Tested across English, Hindi, Hinglish

📊 FX Analytics & Predictive Modeling (SSRN)

Multi-Source FX Analytics and Predictive Modeling
📍 NMIMS University
📄 https://ssrn.com/abstract=5439634

🔍 Highlights:

  • Built full pipeline:
    • Multi-source ingestion
    • LSTM forecasting
    • HMM regime detection
    • Backtesting & portfolio optimization
  • Dataset: ~10 years FX data
  • Focus: reproducibility + real-world trading systems

📌 Status: Working Paper / Preprint

🔧 Tech Stack

Languages
Python | TypeScript | SQL

Frontend
React | Vite | Tailwind CSS

Backend
Flask | FastAPI

AI / ML
TensorFlow | Keras | Scikit-learn | OpenCV

Tools
Docker | Git | Ollama | SQLite

GitHub Analytics

🐍 Contribution Snake

snake gif

Fun Fact

"Don't just predict the future — build systems that shape it."

Pinned Loading

  1. ParkiSense ParkiSense Public

    ParkiSense is an advanced AI-powered application designed to assist in the early detection of Parkinson’s disease through keystroke dynamics, voice analysis, and an AI chatbot. Built using Gradio, …

    Jupyter Notebook 1

  2. Chart-Whisper Chart-Whisper Public

    Chart Whisperer is an AI-powered application that analyzes chart images, extracts data, and lets you ask questions about your charts using natural language.

    TypeScript

  3. Agile-Support Agile-Support Public

    A modern, AI-powered support desk application that combines agile project management with intelligent ticket handling. Built with React, TypeScript, and Ollama integration for AI features.

    TypeScript

  4. Forex-Future Forex-Future Public

    This project is a comprehensive Forex price prediction and analysis platform that combines advanced machine learning models with interactive data visualizations. It provides predictions for major c…

    Jupyter Notebook

  5. Esg-Sector-Fund-Lab Esg-Sector-Fund-Lab Public

    The goal is to evaluate whether an alternative composition (Fund B) can improve ESG profile (and possibly risk/return characteristics) while holding constant the count of positions. Weighting schem…

    TypeScript

  6. Cpp_Monte_Carlo-main Cpp_Monte_Carlo-main Public

    This project integrates a C++ Monte Carlo simulation with a Python dashboard. You can: Adjust simulation parameters from the dashboard Compile and run the C++ code with a single click View results …

    Python