BKD.DEV
Open to Opportunities

Bandhan
Kumar Das

Software Development & IoT Engineer

Computer Science student building real-world systems with Raspberry Pi, OpenCV, TensorFlow Lite, and deep learning — where hardware meets intelligence.

Bandhan Kumar Das
IoT · ML · Vision
Raspberry Pi
TensorFlow Lite
OpenCV
Python
CNN Models
Infosys
Google
IoT Rover
Speech Recognition
Model Compression
SOA University
Raspberry Pi
TensorFlow Lite
OpenCV
Python
CNN Models
Infosys
Google
IoT Rover
Speech Recognition
Model Compression
SOA University

Building at the intersection
of code & hardware

I'm a Computer Science student at ITER, SOA University, Bhubaneswar — passionate about systems that bridge the physical and digital world through IoT, computer vision, and machine learning.

From deploying ML models on edge hardware to controlling servo motors with voice commands, I build things that run in the real world — under real constraints.

Currently pursuing my B.Tech (2023–2027) with a CGPA of 7.41, constantly expanding my toolkit across embedded systems, deep learning, and real-time processing.

3+
Real-World Projects
12x
FPS Improvement
~1s
Voice Response Delay
2
Certifications

What I work with

Languages & Core
Python90%
Java65%
GPIO / PWM Control80%
Machine Learning & Vision
OpenCV85%
TensorFlow Lite78%
CNN / Model Compression72%
Hardware & IoT
Raspberry Pi88%
UART / I2C / SPI70%
Speech Recognition (Vosk/STT)75%

Things I've built

🌿
Plant Disease Classification & IoT Rover

67% field accuracy · Web-based navigation

Trained and deployed an ML model achieving 67% accuracy on field data. Built an IoT-based autonomous rover using Raspberry Pi Zero 2 W with real-time image processing and motor driver control.

Raspberry Pi TFLite OpenCV IoT
✋
Hand Gesture Recognition System

2 FPS → 12 FPS via model compression

Real-time Rock-Paper-Scissors gesture classifier using OpenCV + TensorFlow Lite. Designed custom CNN and achieved a 6x performance boost through model compression techniques.

CNN OpenCV TFLite Optimization
🎙️
Voice-Controlled Robotic Arm

~1 second voice-to-action response

Speech-to-action system using STT APIs converting voice commands into GPIO motor signals. Controlled servo motors via PWM with a response latency of just ~1 second.

Vosk GPIO PWM Servo

Academic Background

B.Tech in Computer Science & Engineering
ITER, Siksha 'O' Anusandhan University, Bhubaneswar
2023 – 2027  ·  CGPA: 7.41
Class XII — CHSE Odisha
Tokyo Advanced H.S. School, Balasore
2023  ·  75%
🤖
Artificial Intelligence Primer
Infosys Springboard
April 2026
☁️
Google Workspace: Bring AI to Work
Google
August 2025

Let's connect

BKD

Open to internships, research collaborations, or just a good conversation about IoT and edge AI. Reach out anytime.