Skip to content

ANIRUDH SURABI

Data Engineer • Chicago, IL • s**************@gmail.com • 773****225 • linkedin.com/••••• • drivetube.ai/•••••

Professional Summary

Data Engineer with 0 years of professional experience and an MS in Computer Science (Loyola University Chicago, 2026). Hands-on experience building reproducible data pipelines, preparing clinical and observational datasets, and deploying ML-ready feature stores. Skilled in Python, SQL, PostgreSQL, AWS, FAISS, and data visualization (Power BI). Seeking an entry-level Data Engineer role to apply data pipeline design, ETL, and cloud-based data infrastructure skills to business and analytics problems.

Technical Skills

Programming Languages: Python,JavaScript
Web Technologies: HTML,CSS
Frameworks and Libraries: TensorFlow,Keras,PyTorch,Scikit-learn,Pandas,NumPy,TF-IDF,HDBSCAN,KMeans,LangChain,React,Material UI,OpenCV,Matplotlib,Seaborn
Databases: SQL,PostgreSQL,SQLAlchemy,ORM,ER diagram design,FAISS,ETL pipelines
Cloud and DevOps: AWS,Jupyter Notebook,Linux,Unix
Data and Analytics: Transfer learning,CNNs,Vision Transformer,Power BI
Tools and Methodologies: Git,GitHub,VS Code,OpenAI API,RAG pipelines,OpenAI embeddings
Parallel & High Performance: CUDA,OpenMP,Numba,GPU-accelerated computing,HPC
Web & Frontend: Framer Motion
Computer Vision & Imaging: Pix2Pix,ResNet,DenseNet,AlexNet,U-Net
Productivity & AI Assistants: GitHub Copilot,ChatGPT,Claude AI

Work Experience

Loyola University Chicago
Chicago, IL
Graduate Research Assistant
March 2025 – August 2025
Academic medical imaging research focused on synthetic image generation and quality assessment to support model development and publication.
Tech Stack: Python, TensorFlow, Keras, Pix2Pix, ResNet, DenseNet, VGG, Linux, Git, Jupyter Notebook
  • Led experimental design and end-to-end reproducible workflows for evaluating physician-rated quality of synthetic medical images, coordinating dataset preprocessing and experiment tracking.
  • Implemented and trained CNN classifiers (AlexNet, ResNet50, DenseNet201, VGG16) on remote Linux GPU servers using transfer learning to predict physician-assigned quality labels.
  • Developed a Pix2Pix image-to-image translation pipeline to synthesize high-fidelity medical images, expanding labeled training data for downstream model experiments.
  • Applied data augmentation, targeted fine-tuning, Dropout, and L2 regularization to reduce overfitting and improve model generalization across validation folds.
  • Designed evaluation and visualization artifacts (confusion matrices, loss/accuracy curves) and translated results into technical summaries for faculty and cross-disciplinary collaborators.
  • Managed dataset curation, annotation verification, and reproducible codebase versioning using Git and Jupyter notebooks to support manuscript preparation.
Rinex Technology
Remote
Machine Learning Intern
March 2022 – April 2022
Instructor-led machine learning certification program delivering applied supervised and unsupervised learning training and hands-on model development.
Tech Stack: Python, Scikit-learn, Pandas, NumPy, HDBSCAN, KMeans, TF-IDF, Jupyter Notebook, Git
  • Completed an intensive ML certification covering supervised and unsupervised algorithms, model evaluation methods, and practical ML workflows.
  • Built end-to-end predictive models using Python, Pandas, and Scikit-learn for classification and regression exercises, applying cross-validation and hyperparameter tuning.
  • Implemented clustering and unsupervised pipelines using KMeans and HDBSCAN and engineered TF-IDF text features for analysis assignments.
  • Evaluated models using precision, recall, F1, and confusion matrices and documented reproducible experiments in Jupyter notebooks.
  • Participated in peer code review and Git-based workflows to iterate on model implementations and ensure code quality during the program.
  • Translated theoretical ML concepts into applied pipelines and reusable notebooks that formed the basis for follow-on academic projects.

Projects

MedReport-QA — Clinical Report Q&A System
Tools Used: Python, LangChain, OpenAI API, FAISS, RAG, OpenAI embeddings
  • Built a retrieval-augmented generation (RAG) pipeline that ingests clinical reports (PDF/TXT), chunks and embeds content with OpenAI embeddings, and serves answers to natural language clinical queries via GPT-3.5-turbo.
  • Designed a custom prompt template and grounding strategy to reduce hallucinations and ensure responses remain strictly within report context for safer clinical Q&A.
  • Integrated FAISS vector search and evaluated retrieval quality to improve answer relevance for clinical information retrieval workflows.
AI-Generated Image Detection
Tools Used: Python, PyTorch, ResNet, Vision Transformer, Scikit-learn
  • Trained and compared ResNet-18 and Vision Transformer (ViT-B/16) models on the CIFAKE dataset for real vs. AI-generated image classification.
  • Achieved strong test accuracy (97.43% for ResNet-18; 98.34% for ViT) and conducted precision, recall, F1, and confusion matrix analysis to validate model robustness.
  • Analyzed architectural differences to demonstrate ViT's benefits in capturing global image features over CNN baselines.
Parallelizing Conway’s Game of Life
Tools Used: Python, NumPy, Numba, OpenMP, CUDA
  • Implemented serial, OpenMP-parallelized, and CUDA GPU-accelerated versions of Conway's Game of Life and benchmarked across grids up to 16,384×16,384.
  • Achieved significant CPU and GPU speedups versus the serial baseline, identified memory bandwidth bottlenecks, and documented performance characteristics across implementations.
Brain Tumor Detection Using Deep Learning
Tools Used: TensorFlow, Keras, ResNet, DenseNet, VGG
  • Developed transfer learning pipelines for binary and multi-class MRI brain tumor classification using ResNet50, DenseNet201, and VGG16 backbones.
  • Applied feature engineering, augmentation, and hyperparameter tuning to achieve consistent test accuracy above 85% across model variants.
Real-Time Pothole Detection System
Tools Used: Python, TensorFlow, Keras, OpenCV, CNNs
  • Built a CNN-based real-time pothole detector with sub-100ms inference latency, achieving high operational accuracy during field tests across varied lighting and road conditions.
  • Integrated detection output into vehicle navigation pipeline and validated performance with empirical field testing.
Scene Understanding in Construction Sites
Tools Used: Python, TensorFlow, Keras, Computer Vision, CNNs
  • Designed and trained a deep learning model to detect construction materials and site objects, achieving high accuracy and average precision in live deployments.
  • Built a worker-facing UI to surface detections and streamline inventory monitoring, reducing material tracking errors in pilot deployments.
Job Market Clustering & Ghost Job Detection
Tools Used: Python, Scikit-learn, HDBSCAN, KMeans, TF-IDF
  • Applied HDBSCAN and KMeans clustering with TF-IDF text features to LinkedIn job posting data to surface hiring demand patterns and potential ghost postings.
  • Built an interactive dashboard and validated cluster quality using silhouette and lift metrics to provide actionable hiring trend insights.
Restaurant Management System
Tools Used: PostgreSQL, SQLAlchemy, Python, ORM
  • Designed a normalized relational database schema with referential integrity constraints for order, customer, and feedback workflows.
  • Implemented complex SQL queries and ORM models to ensure end-to-end data consistency for restaurant operations.
Gamified Productivity Tracker
Tools Used: React, JavaScript, Framer Motion, Material UI, GitHub
  • Developed a React-based task tracker with animated progress visuals and badge mechanics to increase user engagement.
  • Led UI debugging, form validation, and collaborative GitHub workflows for deployment across the full development lifecycle.

Education

Loyola University Chicago
Master of Science in Computer Science • Chicago, IL • May 2026
Gokaraju Rangaraju Institute of Engineering and Technology
Bachelor of Technology in Information Technology • India • August 2023

Certifications

Python Programming Certification — HackerRank
Machine Learning Certification — Rinex Technology
Python Computing Course — NPTEL / IIT
Artificial Intelligence Training — Oracle Academy
Machine Learning Session — Cognizance, IIT Roorkee
DevOps Workshop — IIT Varanasi
MIT App Design Workshop — IETE
User-Centric Computing for Human-Computer Interaction

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