Hi, I'm Souri Rishik

Full-Stack Developer

I'm a Data Science Engineering student at MIT, Manipal, and a passionate software developer who thrives on bringing ideas to life through code. Driven by a deep curiosity for technology, I am constantly exploring new challenges to learn, grow, and build impactful solutions.

View Projects

Download Resume

About Me

A curious mind at the intersection of data science, robotics, and software engineering — always learning, always building.

Education

Manipal Institute of Technology, MAHE

Jul 2023 — Sep 2027

B.Tech in Data Science Engineering

Mahatma Gandhi Memorial PU College, Udupi

Jul 2021 — Apr 2023

Higher Secondary Education (PCMC) — Karnataka State Board

Little Rock Indian School, Brahmavar

Mar 2021

CBSE 10th Board

Experience

Coding Subsystem Head

Sep 2025 — Present

Robomanipal

Enhanced the official Robomanipal website with 100%+ traffic growth and 98/100 Real Experience Score. Represented MIT MAHE at Technoxian 8.0 (quarterfinals among 200+ teams). Secured 2nd Runner-Up at Robonautica, IISc Bangalore among 179 teams nationwide.

Research Intern

Jun 2025 — Jul 2025

Manipal Institute of Technology

Engineered ML models (Gradient Boosting, CatBoost, XGBoost, Random Forest) to detect phishing URLs with 97.4% accuracy and 97.7% F1-score on 30,000+ samples. Designed and deployed a Python-based phishing detection web app with real-time URL classification, reducing manual verification time by 70%.

Skills & Technologies

Python

C/C++

Java

JavaScript

TypeScript

HTML

CSS

React

Next.js

TensorFlow

PyTorch

Python

C/C++

Java

JavaScript

TypeScript

HTML

CSS

React

Next.js

TensorFlow

PyTorch

NumPy

Pandas

OpenCV

Git

GitHub

AWS

VS Code

Ubuntu

SQL

Keras

ROS

NumPy

Pandas

OpenCV

Git

GitHub

AWS

VS Code

Ubuntu

SQL

Keras

ROS

Certifications

DEEPLEARNING.AI · COURSERA

Generative AI with Large Language Models

A comprehensive deep dive into the generative AI lifecycle. Covers transformer architecture, LLM training, fine-tuning, and deployment.

Verify

DEEPLEARNING.AI · COURSERA

Natural Language Processing in TensorFlow

Explores advanced NLP techniques including tokenization, sentiment analysis, word embeddings, RNNs, and LSTMs using TensorFlow.

Verify

COURSERA

MongoDB: The Complete Guide to NoSQL Database Development

Master MongoDB NoSQL database development, covering CRUD operations, indexing, aggregation framework, and data modeling.

Verify

COURSERA

C++ Data Structures in the STL

An in-depth exploration of C++ Standard Template Library, focusing on vectors, maps, sets, and efficient algorithmic implementations.

Verify

COURSERA PROJECT NETWORK

Build Your First GUI App With Java

Hands-on project developing a graphical user interface application using Java Swing and basic event-driven programming.

Verify

COURSERA

Introduction to Java Programming: Java Fundamental Concepts

Foundational concepts of Java programming including object-oriented design, syntax, control structures, and basic data types.

Verify

My Projects

A collection of projects spanning robotics, AI, computer vision, and web development. Each one taught me something new.

Sign Language Gesture Analysis

Built a lightweight custom CNN (~227K parameters) to classify 24 static ASL hand gestures from the Sign Language MNIST dataset. Achieved 99.69% validation accuracy and demonstrated real-time webcam inference, correctly recognizing 20/24 gestures under live conditions.

PythonTensorFlowKerasOpenCVNumPy

View on GitHub

Phishing URL Detection

Engineered multiple ML models (Gradient Boosting, CatBoost, XGBoost, Random Forest) to detect phishing URLs, achieving 97.4% accuracy and 97.7% F1-score on 30,000+ samples. Deployed a real-time Python web application reducing manual verification by 70%.

PythonXGBoostCatBoostMachine Learning

View on GitHub

Digit Recognition

Developed a Feedforward Neural Network in TensorFlow/Keras with preprocessing, batch normalization, dropout, and L2 regularization for MNIST digit recognition. Achieved 98.13% classification accuracy.

PythonTensorFlowKerasNumPyPandas

View on GitHub

Fastest Line Follower

Optimized PID algorithm with manual, Ziegler–Nichols, and incremental tuning, improving motor precision by 30%. Created a Bluetooth-based Python GUI for real-time PID tuning, cutting adjustment time by 50%. Competed at Technoxian 8.0, advancing to quarterfinals among 200+ teams.

CPythonSTM32CubeIDEPIDEmbedded

View on GitHub

Autonav (RoboCUP@Work)

Programmed a 6-DOF Mecanum robot for RoboCUP@Work with 2D LiDAR and depth camera, using SLAM Toolbox and NAV2 for mapping and navigation. Achieved <10cm localization error and 92% goal success in ROS2.

ROS2SLAMLiDARNAV2Robotics

View on GitHub

Open Volley

Built a computer vision system in Python/OpenCV to track volleyball trajectories with 95% detection accuracy, processing 120+ frames per second for real-time analysis. Automated ball tracking and player count per team, reducing manual annotation time by 60%.

PythonOpenCVComputer Vision

View on GitHub

Get In Touch

Have a question, want to collaborate, or just want to say hi? Drop me a message and I'll get back to you soon.

souri.rishik27@gmail.com

GitHub

github.com/SouriRishik

linkedin.com/in/souri-rishik

Always open to interesting conversations and opportunities!