Indu.

Hi, I'm Indumathi šŸ‘‹

About Me

I’m a Computer Science graduate student at the University at Buffalo (SUNY) specializing in Artificial Intelligence and Machine Learning. I enjoy building intelligent systems that work reliably in real-world, real-time settings.

My core interests lie in Deep Learning and Computer Vision, where I have worked on projects such as elder-care fall detection using pose estimation and aerial object detection for aircraft and drones. I focus not just on model accuracy, but also on evaluation, robustness, and deployment constraints.

Prior to graduate school, I worked as a Data Engineer at Accenture, where I built and optimized data pipelines, ensured data quality, and worked with cloud platforms like BigQuery and Google Cloud Storage. This experience helps me approach ML problems end-to-end — from clean data to reliable models.

I’m actively seeking AI/ML Engineer, Computer Vision Engineer, or Data Engineer roles where I can contribute to impactful, production-ready systems and continue learning.

My Toolkit

Python
Java
MySQL
PyTorch
TensorFlow
Docker
Google Cloud
GitHub
Linux

Programming Languages

Core languages I use for building systems.

Programming Languages

  • Python
  • Java
  • C / C++

Web

Frontend fundamentals for portfolio + apps.

Web

  • HTML
  • CSS
  • JavaScript

Database

Querying, schema design, and analytics.

Database

  • SQL
  • MySQL
  • BigQuery

ML / Data

Training, evaluation, and pipelines.

ML / Data

  • PyTorch, TensorFlow
  • Precision / Recall / F1
  • ETL, Data Pipelines
  • Docker, GCP

Projects

Coursework and hands-on projects from my MS in Computer Science (SUNY Buffalo).

Student Budget & Expense Management System

Student Budget & Expense Management System

Student Budget & Expense Management System

Designed a normalized relational DB for student expense tracking with BCNF, constraints, and analytics queries.

SQLBCNFER Modeling
GitHub
AI-Powered Elder Care Monitoring System

AI-Powered Elder Care Monitoring

AI-Powered Elder Care Monitoring

ResNet18 + LSTM activity recognition with fall detection + email alerts. Achieved 79.04% test accuracy.

ResNet18+LSTMHMDB51Fall Detection
GitHub
Autoencoders for Anomaly Detection

Autoencoders for Anomaly Detection

Autoencoders for Anomaly Detection

Unsupervised anomaly detection using reconstruction error on sensor time-series.

AutoencoderTime-SeriesUnsupervised
GitHub
Transformer from Scratch in PyTorch

Transformer from Scratch (PyTorch)

Transformer from Scratch (PyTorch)

Implemented attention + positional encodings + training loop for text classification.

TransformersAttentionYelp Polarity
GitHub
Summarization using LLMs (BillSum)

LLM Summarization (BillSum)

LLM Summarization (BillSum)

Abstractive summarization pipeline for long U.S. bills to concise summaries.

NLPSummarizationBillSum
GitHub
Time-Series Forecasting using RNNs

Time-Series Forecasting (RNNs)

Time-Series Forecasting (RNNs)

RNN/LSTM forecasting on household power consumption sequences.

RNN/LSTMForecastingTime-Series
GitHub
Sentiment Analysis using LSTM

Sentiment Analysis (LSTM)

Sentiment Analysis (LSTM)

LSTM classifier for airline tweet sentiment (pos/neg/neutral).

LSTMText ClassifierKaggle
GitHub
Panorama Stitching & Image Alignment

Panorama Stitching & Alignment

Panorama Stitching & Alignment

Feature matching + homography + blending to build panoramic outputs.

Feature MatchingHomographyBlending
GitHub

Get in Touch!

I'm always excited to hear about new opportunities and collaborations. Don't hesitate to reach out and let's make something great.

Contact Me