Vishal Pandey | ML Research Engineer; Neuroscience

🚀 Projects

Here are some of the projects I’ve built, spanning AI research, applied ML, systems, and real-world use cases.

Research & Real-World Projects

Project Description Link(s) / Notes
GenAI Research Agent An autonomous research assistant for literature review and knowledge extraction. Uses LLMs to explore papers, summarize findings, and suggest new directions. GitHub
Neuromorphic Decision Transformer (arXiv Paper) Spiking neural variant of Decision Transformers to reduce inference energy while retaining policy performance. GitHub
SmartTraffic-RL Reinforcement learning framework for adaptive urban traffic signal control. Includes gym-style RL environment and integration with SUMO. GitHub
Repo Ingestion Agent Agentic tool that ingests GitHub repos, parses code/documentation, structures them for use in downstream LLM tasks. GitHub
KibouChan Chatbot focused on mental wellness & positive interactions. Blends conversational flow control with LLMs. GitHub
AgroSense (arXiv Paper) Crop recommendation system leveraging soil and climate data; built for precision agriculture. GitHub
AgenticCyberOps Autonomous cybersecurity pipeline: penetration testing, threat detection, automated response. GitHub
NumPyLiteC Lightweight C-based implementation of NumPy-like operations; focusing on speed and minimal footprint. GitHub
PDF2Podcast Converts research papers / PDFs into podcasts via summarization + text-to-speech. GitHub
pandas.c C re-implementation of core Pandas features; exploring efficiency and systems level trade-offs. GitHub
Handwritten Recognition (MNIST) Deep learning model for digit recognition using the MNIST dataset. GitHub
Wave Function Collapse Procedural content generation using the Wave Function Collapse algorithm. GitHub
Stay Productive Productivity app to track tasks, manage focus sessions, encourage consistent workflow. GitHub
Command-Line Game Interactive command-line game; lightweight, fun and demonstrates UI/UX within terminal. GitHub

Kaggle & Applied ML Projects

Project Description Link(s) / Notes
Signature Verification (CNN, SNN, CSNN) Verifies signatures using Convolutional Neural Networks, Spiking Neural Networks, and Continuous Spiking Neural Networks. Kaggle
Drowsiness Detection Uses data/image inputs to detect driver’s drowsiness; useful for safety systems in real time. Kaggle
Malware Detection (DT, SVM, LR, NN) Comparative study of classification methods (Decision Trees, Support Vector Machines, Logistic Regression, Neural Nets) for detecting malware. Kaggle
Signature Verification (RNN / C-RNN / B-RNN) Uses various recurrent neural network architectures (vanilla RNN, conditional RNN, bidirectional RNN) for verifying signatures. Kaggle
Real Forge Signature Detection Focuses on distinguishing forged signatures vs real signatures; uses image analysis. Kaggle
Evaluate OpenLLMs for Text Summarization Benchmarks open-source LLMs on summarization tasks; compares quality vs inference cost. Kaggle
Diabetes Prediction and Its Types (Bachelor’s Project) Predicts various types of diabetes and explores feature importance; includes data preprocessing, model comparison and interpretation. Kaggle

Publications & Highlights