Skip to content

Sai Ganesh Devi Prasad Robbi

AI/ML Engineer • r**************************@gmail.com • 990****976 • drivetube.ai/•••••

Professional Summary

AI/ML Engineer with 0 years of experience building and deploying computer vision and medical imaging models using FastViT, Python and Flask. Strong foundations in data structures, OOP and deep learning with hands-on project experience in X-ray and CT lung disease detection.

Technical Skills

Programming Languages: Python,Java,C,C++
Frameworks and Libraries: PyTorch,FastViT,Flask,scikit-learn,NumPy,OpenCV,Pandas,Matplotlib
Data and Analytics: Machine learning,Tableau
Tools and Methodologies: Git,GitHub,VSCode,Reproducible experiment tracking
Computer Vision & Imaging: Image preprocessing,Data augmentation,Intensity windowing,Slice selection
ML Modeling & Evaluation: Deep learning,Hyperparameter tuning,Cross-validation,Stratified sampling,Model evaluation Accuracy, Precision, Recall, F1-score

Work Experience

BSNL
Intern
SEPT 2024 – OCT 2024
Interned at BSNL, India's state-owned telecommunications provider; supported engineering tasks related to network data processing and reporting.
Tech Stack: Python, Git, GitHub, Tableau, VSCode, Pandas
  • Developed Python scripts to parse and preprocess telecom log and performance data to streamline downstream analysis by engineers.
  • Applied data structures and algorithmic techniques to optimize preprocessing routines for memory and runtime efficiency.
  • Built Tableau visualizations to present network KPIs and assisted in preparing reports used by the engineering team.
  • Documented standard data formats and processing steps to improve reproducibility and handover for subsequent analysts.
  • Assisted in testing and validating automation scripts across sample datasets to ensure correctness and robustness.
  • Managed code and version control using GitHub; maintained readable repository structure and README documentation.
Centre of Excellence in Maritime & Shipbuilding
Intern
MAY 2025 – JUN 2025
Interned at a maritime and shipbuilding technology center focused on research and prototype development; supported imagery and data-driven prototype tasks.
Tech Stack: Python, FastViT, Flask, PyTorch, OpenCV, GitHub
  • Implemented Python-based data processing pipelines for imagery and sensor datasets to prepare inputs for vision model experiments.
  • Applied image preprocessing, intensity windowing and data augmentation techniques to improve model training robustness for imagery tasks.
  • Prototyped FastViT-based vision model experiments, conducting hyperparameter tuning and cross-validation to measure model performance.
  • Built a Flask-based prototype interface to enable model inference and visualization for demonstration and stakeholder review.
  • Logged experiment results and maintained model checkpoints and reproducible training scripts in the project repository.
  • Presented experimental findings and technical documentation to mentors to support decision-making on model selection.

Projects

MedScanXR – Lung X-ray Disease Detection
Tools Used: Python, FastViT, PyTorch, Flask, OpenCV, NumPy, Data augmentation, Stratified sampling, Model evaluation
  • Designed and trained a FastViT-based deep learning model to detect lung diseases from chest X-ray images, focusing on maximizing recall and F1-score.
  • Implemented advanced image preprocessing including resizing, normalization and augmentation pipelines to improve model generalization.
  • Used stratified data splitting to maintain class balance and prevent biased model training during evaluation.
  • Conducted hyperparameter tuning and evaluated models using Accuracy, Precision, Recall and F1-score to select the best-performing checkpoint.
  • Deployed the trained model as a Flask web application enabling real-time image upload and disease prediction through REST endpoints.
MedScanCT – Lung CT Scan Analysis
Tools Used: Python, FastViT, PyTorch, Flask, Intensity windowing, Slice selection, Cross-validation, Model evaluation
  • Built a FastViT-based model tailored for lung disease detection from CT scan slices, optimizing for robust feature extraction across slices.
  • Applied domain-specific preprocessing such as intensity windowing, normalization and slice selection to enhance relevant feature contrast.
  • Employed patient-wise and stratified splitting strategies to prevent data leakage and ensure robust evaluation.
  • Validated model performance through cross-validation and tracked metrics to compare model variants.
  • Developed a Flask-based web interface for real-time CT scan analysis and visualization of model predictions.

Education

Gitam Deemed to be University
B.Tech ECE (AIML) • AUG 2022 – APR 2026
Sri Chaitanya Junior College
MPC • MAR 2020 – APR 2022
Sri Chaitanya EM School
SSC • MAR 2019 – MAR 2020

Achievements

  • Solved 260+ DSA problems across LeetCode and GeeksforGeeks

Powered by Drivetube · Create your own profile at drivetube.ai