👋 Hello World, I'm

PAVAN B

Building cool things through code
AI Full Stack Applications Python
Pavan B

Passionate Developer | AI Enthusiast | Innovator

Hello! I'm Pavan, a Software Engineer with a strong passion for System Design, Machine Learning, and Full Stack Applications.

Currently, I'm an Associate Software Engineer at TexasAI, working on Artificial Intelligence building incredible things :) I have also interned at Mipzo Pvt Ltd, Silo Fortune and Akanksha Charitable Trust, gaining experience in AI and software development.

I have completed my engineering through MCE in Information Science on 2025. I served as the Vice President of The Malnad Technical Club and actively participate in NSS, fostering teamwork, leadership, and technical excellence.

Let's connect and collaborate on exciting projects!

My Projects

AgroIntent Screenshot

AgroIntent — AI Crop Assistant

AI-powered crop diagnostic tool built for PromptWars Hackathon (Google for Developers + H2S).

Features:

  • Vertex AI (Gemini 2.5 Flash) for crop diagnosis
  • Serverless FastAPI on Google Cloud Run
  • Voice I/O in 5 languages via Web Speech API
  • Urgency level + AI confidence scoring
Video Conferencing App Screenshot

NeuroSpectra

A full stack application for predicting the certainity of a person having autism based on A0 to A10 scores.

Features:

  • ASD prediction accuracy 99.98%
  • SVM vs RF vs LR benchmarking
  • Backend model inferencing endpoint built with fastAPI
  • Deployed model artifacts on render
Video Conferencing App Screenshot

Leo — Ride Booking Platform

Full-stack ride booking platform with Driver, Rider, and Admin apps. Built as a consulting project.

Features:

  • React Native + Spring Boot + MongoDB
  • Google Maps for geolocation & route tracking
  • Twilio for real-time ride notifications
  • Razorpay for secure payment processing
  • Microservices deployed on AWS
Video Conferencing App Screenshot

Video Conferencing App

A group video chatting and live streaming application using Agora RTC.

Features:

  • Real-time video/audio streaming
  • Screen sharing capability
  • Group Chat feature
  • Low latency communication
Handwritten Digit Recognition Screenshot

Handwritten Digit Decoder

TensorFlow-based model for handwritten digit recognition with 98% accuracy

Technical Details:

  • ANN architecture with TensorFlow
  • 98% accuracy on MNIST dataset
  • Real-time prediction capability
  • Custom preprocessing pipeline

View All Projects

Explore my complete portfolio of projects and contributions on GitHub.

On My GitHub:

  • More coding projects and games
  • Open source contributions
  • Code repositories
  • Development history
Click to visit my GitHub profile →

Skills Matrix

Experience Timeline

TexasAI

Mar 2025 - Present

Associate Software Engineer

Working on development of backend systems with FastAPI. Integrating AI models to provide real-time solutions.

Mipzo

Nov 2024 - Feb 2025

Innovation Technical Intern

Led LMS development and implemented innovative frontend solutions using modern technologies.

Malnad Technical Club

Jan 2022 - Dec 2024

Vice President & Head of Technical Committee

Organized technical events and led team initiatives, fostering innovation and technical excellence.

Certifications

AWS Certified Machine Learning Engineer

Apr 2026

Associate — Amazon Web Services

Designed and deployed end-to-end ML pipelines on AWS — spanning data preparation, model training with SageMaker, inference via Bedrock, and monitoring.

Score: 814/1000 SageMaker Bedrock MLOps
Hugging Face Logo

LLM Course — Fine-tuning Language Models

Dec 2025

Hugging Face

Fine-tuning language models — covering supervised fine-tuning, instruction tuning, and evaluation of LLMs.

Fine-tuning LLMs Transformers

Get in Touch