Abhin Krishna

Kochi, Kerala, India | dearabhin [at] gmail [dot] com

About Me

I am an Electronics and Biomedical Engineering student at Model Engineering College, Kochi, with a deep-seated interest in the convergence of Artificial Intelligence, Deep Learning, and Neuroscience. My work focuses on Brain-Computer Interfaces (BCI) and Neurotechnology, where I apply my skills to develop assistive technologies and enhance medical diagnostics. I am continuously strengthening my foundation in engineering mathematics and AI research methodologies while actively exploring EEG-based BCI systems for decoding neural signals. As an active member of technical communities, I thrive on collaboration and am dedicated to contributing to impactful applications in healthcare and human-machine interaction.

Education

Bachelor of Technology - B.Tech, Electronics and Biomedical Engineering

Model Engineering College, Thrikkakara (KTU)

September 2024 – Present

Skills

Web Development API Machine Learning Deep Learning Computer Vision Object Detection Natural Language Processing (NLP) Large Language Models (LLM) TinyML Research CAD Data Analysis & Visualization Brain-Computer Interfaces (BCI) EEG/EMG Signal Processing MATLAB Python R C C++ HTML & CSS JavaScript Arduino ESP32 Firmware Development Bluetooth Low Energy (BLE) Git & GitHub Linux Microsoft Office DNS Cloud Services (AWS, GCP)

Projects

Decoding Imagined Handwriting from EEG

Aug 2025 - Present

A machine learning project focused on classifying high-dimensional, noisy EEG time-series data to decode imagined handwriting for BCI applications. This involves preprocessing 32-channel EEG signals and implementing EEGNet, a compact CNN, to perform a four-way classification of imagined letters ('L', 'V', 'O', 'W'). The project investigates the model's robustness to temporal variance and the challenges of low signal-to-noise ratio in single-trial EEG.

BCI EEG Machine Learning CNN Signal Processing

Real-Time Object Detection for Edge Devices using TinyML

Aug 2025 - Present

Developed a real-time, multi-label object detection and classification system on a resource-constrained microcontroller (Seeed Studio XIAO ESP32S3 Sense). The system identifies and differentiates between "Cats" and "Dogs" at ~7 FPS with 143 ms latency, demonstrating a complete TinyML workflow from data collection to firmware deployment using Edge Impulse Studio and the FOMO algorithm.

TinyML Edge AI Computer Vision Object Detection ESP32 Arduino C++

Bitchat CLI: A Decentralized Python P2P Messenger

Jul 2025 - Present

Engineered a Python-based CLI for a secure, decentralized P2P messenger running over a Bluetooth LE mesh network. Inspired by Jack Dorsey's original project, it brings a serverless chat experience to the command line with end-to-end encrypted messaging and public channels.

Python Bluetooth Low Energy (BLE) P2P Networking Cryptography

LipiPala: Preserving India's Linguistic Heritage

Feb 2025 - Present

An open-source initiative to preserve, document, and revitalize endangered Indian languages using AI. The project combines advanced NLP, speech recognition, and community collaboration to create accessible tools for indigenous communities.

Natural Language Processing (NLP) Large Language Models (LLM) Speech Recognition Open Source

Multimodal AI Medical Agent

Jan 2025 - Apr 2025

Developed a multimodal AI medical agent utilizing the capabilities of the Gemini 2.0 model to process and interpret various data types for medical applications.

Python Generative AI Multimodal AI

EMG based Smart Home Automation

Dec 2024 - Feb 2025

A research-oriented project focused on using Electromyography (EMG) signals for smart home automation. The project involved machine learning for signal classification, electronics for hardware interfacing, and overall project management.

Electromyography (EMG) Machine Learning Electronics IoT

Web Screenshot Bot - Telegram Mini App

Jul 2022 - Dec 2024

Created a Telegram mini-app to safely take screenshots of webpages from URLs without opening potentially risky links directly on a device, enhancing user security.

JavaScript Python Telegram API Web Scraping

Research Interests

  • Brain-Computer Interfaces: EEG/EMG for assistive technologies and neural decoding.
  • AI in Healthcare: Generative AI (e.g., Imagen) and Deep Learning for diagnostics and personalized medicine.
  • Machine Learning: Efficient Deep Learning (TinyML), Reinforcement Learning, and NLP.
  • Neurotechnology: Developing novel hardware and software for interfacing with the nervous system.

Blog Posts

A guide to AI and ML

Artificial intelligence (AI) and machine learning (ML) are changing the future of work.

Read More

Sympy Guide

SymPy is a Python library for symbolic mathematics, allowing manipulation of mathematical expressions programmatically.

Read More