Open to Opportunities

Mrudula Gharat

I am an AI & Machine Learning Developer focused on deep learning, intelligent healthcare solutions, and scalable architectures. Bridging the gap between cutting-edge research and production systems.

View Work GitHub
01

Background

As a B.Tech student specializing in Artificial Intelligence and Machine Learning, I combine a rigorous theoretical foundation with hands-on engineering practice.

My expertise spans developing conversational AI, constructing deep learning models, and building robust web interfaces to make AI models accessible to end-users.

developer.py bash — ai_agent_output.log 
import torch
from models import ResNet, Attention

class AIDeveloper:
    def __init__(self):
        self.focus = ["Deep Learning", "CV", "NLP"]
        self.stack = ["PyTorch", "Python", "React"]
        
    def build(self, problem):
        # Define robust architecture
        model = Attention(ResNet())
        return model.solve(problem)

>>> Initializing Neural Core... [OK]
>>> Loading pre-trained weights (ResNet-152)... [OK]
>>> Running inference on input payload...

[OUTPUT]
Class ID: 0x01 (Damage Detected)
Confidence Score: 98.74%
Bounding Box: [102, 45, 340, 280]
Latency: 12ms (CUDA accelerated)

$ Awaiting user confirmation_|
02

Expertise

Machine Learning

Designing predictive models, deep neural networks (ResNet, CNNs), and handling complex data pipelines for computer vision and NLP.

PyTorch TensorFlow Scikit-Learn

Software Engineering

Writing clean, efficient, and scalable code in Python and JavaScript. Building APIs, integrating models into production systems.

Python SQL Git

Full-Stack Development

Creating seamless user interfaces and robust backends to deploy machine learning applications to the interactive web environments.

React Node.js WebSockets
03

Work Experience

AI Intern @ Rams

April 2025 – Present

Ongoing

Developing an AI-powered Rack Damage Detection System using computer vision.

🚀 Project: Built a multi-class object detection model (YOLOv5) to identify Damage, Defects, Arrangement Issues, and Protector Issues.
  • Annotated and meticulously managed a dataset of 3000+ images.
  • Trained and optimized deep learning models using PyTorch.
  • Improved model performance through dataset balancing & hyperparameter tuning.
  • Evaluated results using robust metrics including Precision, Recall, and mAP.
  • Utilized Docker for streamlined development and deployment.
Python PyTorch YOLOv5 OpenCV Docker CUDA
04

Selected Work

05

Get In Touch

Currently seeking opportunities where I can apply machine learning to solve challenging problems. Whether you have a question or just want to say hi, I'll try my best to get back to you!

Say Hello