Hello, my name is
Yashaswini Madineni
I build at the intersection of AI and software engineering, creating solutions that scale and think. I'm a Master's student in Applied Data Science at San Jose State University, and I bring 2 years of experience as a Software Engineer working on ETL pipelines, distributed systems, and cloud-native applications. My projects focus on Generative AI, Agentic AI, LLMs, deep learning, scalable ML systems, and Big Data workflows, where I've combined AI and data engineering to solve real-world challenges.
I'm passionate about building solutions that bring together AI and software systems, and I'm actively seeking full-time opportunities in Software Engineering (AI/ML), AI Engineering, and Data Engineering.
When I'm not building AI agents, I'm acting like one by automating job applications at scale!
Computer Science
Bachelor of Engineering
2018–2022
Courses
- Data Structures & Algorithms
- Operating Systems
- Advanced Computer Networks
- Software Engineering
- Object-Oriented Programming
- Computer Architecture
- Compiler Design
Data Science Intern
Tequed Labs
Jan 2022 - May 2022
Work highlights
- Developed and evaluated ML models (Logistic Regression, Random Forest, SVM) using Scikit-learn, improving predictive accuracy on structured datasets by 15%.
- Implemented deep learning experiments in PyTorch and TensorFlow, building and testing neural networks for classification tasks.
- Built analytics workflows in Python (Pandas, NumPy, PySpark) to preprocess and engineer features from large-scale datasets, reducing data preparation time by 20%.
- Conducted exploratory data analysis (EDA) and created dashboards with Tableau and Power BI, enabling stakeholders to identify key business trends
Software Engineer
Tata Consultancy Services
2022–2023
Work highlights
- Designed and maintained scalable ETL pipelines using Python, SQL, Snowflake, and AWS (Redshift, Glue, S3, EMR), improving data flow efficiency by 20% through A/B testing.
- Engineered batch and streaming data pipelines with Apache Spark, Hadoop, and BigQuery, optimizing query execution for large-scale analytics and reducing query latency by 30%.
- Containerized and orchestrated data workflows with Docker and Kubernetes, enabling reproducible deployments and improving scalability and fault tolerance in production environments.
- Built interactive dashboards with Tableau to deliver data-driven insights, collaborating with product managers, data scientists, and platform engineers to translate requirements into scalable solutions.
Masters in Applied Data Science
San Jose State University
2023–2025
Courses
- Generative AI
- Deep Learning
- Big Data Technologies and Applications
- Machine Learning Technologies
- Distributed Systems
- Database Systems
- Data Visualization
- Mathematical Methods for Data Science
Things I've Built
Featured Project · In Progress
AI Mock Interview Practice Platform
An AI-powered mock interview platform for Data Science that generates dynamic, domain-specific questions, evaluates responses, and delivers instant feedback across SQL, Python, Statistics, Machine Learning and more, enhancing interview preparation. Built end-to-end with cloud-native data pipelines, NLP, and Generative AI, the platform showcases expertise in scalable system design, applied machine learning, and AI engineering to help learners prepare for real-world technical interviews.
LLM-Powered Text-to-SQL with AI Agents
A GenAI pipeline that translates natural language into SQL queries using fine-tuned LLaMA-3 and Mistral on the Spider dataset. Integrated LangChain agents (generation, clarification with Gemini, optimization with OpenAI GPT-3.5), and deployed with a Flask backend and HTML/CSS/JS frontend. Model hosted on Hugging Face with Docker & CI/CD support.
Agentic AI Systems: MultiModal RAG, Stable Diffusion Fine-Tuning & Travel Assistant
Developed an end-to-end AI system combining multi-modal RAG, generative model fine-tuning, and an agentic AI travel assistant. The RAG system retrieved text & images from PDFs and used Gemini for grounded Q&A, Stable Diffusion was fine-tuned with LoRA for custom image generation, and a multi-agent assistant (Flight, Weather, Hotel, Itinerary) orchestrated APIs to build dynamic travel plans.
Substance Use Prediction & Pattern Analysis using Machine Learning
Designed an end-to-end Machine learning pipeline for analyzing and predicting substance use patterns. Conducted extensive data preprocessing and exploratory data analysis (EDA), applied PCA for dimensionality reduction, KMeans clustering for unsupervised insights, and multiple classifiers (Random Forest, Logistic Regression, Decision Trees, KNN) for supervised prediction. Addressed class imbalance with ADASYN to improve minority class recall.
Crypto Market Analytics with Big Data
Analyzed 48GB of Binance cryptocurrency trading data (2020–2024) using AWS EC2, Databricks, Apache Spark, Pandas, and PyArrow. Built scalable pipelines for data cleaning, preprocessing, and transformation, followed by volatility analysis, trading patterns, and liquidity insights. Developed interactive Tableau dashboards for visualization, showcasing distributed systems, cloud deployment, and big data engineering for financial analytics.
MediPredict — AI Full-Stack Healthcare Prediction System
Built a full-stack distributed healthcare application for symptom-based disease prediction. Integrated a React frontend with a Flask backend, multiple ML models, and secure cloud deployment on GCP. The system supports authentication, interactive consultations, and automated health report generation.
Deep Learning for Skin Cancer Segmentation & Classification (ViT-UNet)
Implemented a hybrid Vision Transformer (ViT) + U-Net model for automated skin cancer analysis. The system performs both segmentation (pixel-level lesion masks) and classification (benign vs. malignant), trained on ISIC dermoscopic datasets. Combined BCE + Dice loss for segmentation and weighted cross-entropy for classification to handle class imbalance, achieving strong performance on ISIC 2016 & 2017 benchmarks.
Get In Touch
If you would like to work together or discuss an opportunity for work, feel free to reach out at yashaswini.madineni@sjsu.edu or connect with me on LinkedIn.