μ
θ
π
θ
θ
Ω
σ
θ
σ
θ
Pranav Kuramkote Sudhir profile picture
H
My LinkedIn profile My Github profile

Get To Know More

About Me

My Background

I'm a software Engineer with 4 years of experience, currently pursuing my masters in Data Analytics Engineering at Northeastern University, Boston. My expertise lies in designing and implementing scalable distributed systems and developing machine learning solutions. With Python as my primary language, I've built and deployed various data-driven applications and ML models. I have extensive experience in cloud computing and containerization, enabling me to deliver robust and scalable solutions. My technical skills span across modern machine learning frameworks, cloud platforms, and data processing technologies.

Scripts

Skill 1 Skill 2 React Node.js Skill 4 Skill 5

Tools

Skill 6 Skill 7 Skill 8 Skill 9 Skill 10

DevOps

Skill 12 Azure GCP Skill 13 Skill 15 Skill 16 Skill 14 Skill 14 Skill 14
Arrow icon

Explore My

Experience

Trainee Engineer

Quest Global Engineering

Nov 2019 - Oct 2020

Bangalore, India

Member Technical Staff

HCL Technologies

Feb 2021 - Jul 2023

Bangalore, India

Manufacturing Data Science Co-Op

Veeco Instrument

Jul 2024 - current

San Jose, California

Arrow icon

Browse My Recent

Projects

Toxic Comment Analyzer

Toxic Comment Analyzer

Built a Natural Language Processing (NLP) model to detect and classify toxic comments in social media discussions. Preprocessed text data using TF-IDF, word embeddings, and tokenization techniques to improve model accuracy. Trained multiple models including Logistic Regression, Random Forest, and LSTM, achieving high precision and recall scores. Developed a Flask-based API to integrate the model into web applications for real-time comment analysis. Deployed on AWS Lambda for cost-efficient, scalable inference.

classification model

Mail classifier

I built a custom GPT-style transformer from scratch using only NumPy and core Python. It includes multi-head attention, layer normalization, and greedy decoding logic. I trained it on email data and evaluated the results with custom metrics. The model can generate replies and summaries for real emails. This project deepened my understanding of LLMs beyond library abstraction.

Personal Chatbot

Rag based chatbot to answer recruiter questions

I built a serverless AI-powered chatbot using React, Node.js, and Tailwind CSS. It leverages Cohere's APIs for both semantic search (embeddings) and natural language generation. PostgreSQL with pgvector on Render is used to store and retrieve relevant context. The backend is deployed as serverless functions via Vercel. This chatbot can answer questions about my experience, skills, and projects in real-time.

Portfolio Website

Portfolio Website

A modern, responsive personal portfolio website designed to showcase my professional journey, skills, and projects. Built using HTML, CSS, JavaScript, and Bootstrap, it features a clean UI with smooth animations and interactive sections. The site includes a dynamically generated projects section, contact form, and responsive navigation for seamless user experience. Implemented GitHub Actions for CI/CD to automate deployment. Hosted on GitHub Pages, ensuring reliability and scalability.

Get in Touch

Contact Me

pranavkuramkotesudhir@gmail.com