
I'm Aishwarya Subrahmanya

Name: Aishwarya Subrahmanya
Profile: Data Engineer/ Data Analyst
Email:
belakavadisubrahma.a@northeastern.edu
aishwaryabs6@gmail.com
Phone: (857) 396-4932
About me
I am a graduate of Northeastern University with a Master’s in Data Analytics Engineering and I hold a B.Tech in Computer Science from PES University. I work with data end-to-end cleaning, processing and transforming datasets, building ETL pipelines, modeling databases and developing dashboards and reports that support decision-making.
Strong in Python and SQL and proficient with Tableau, Excel, Google Sheets, Spark, and Statistics, I also use AWS and Snowflake for scalable data solutions. I am continuously upskilling in machine learning, staying curious and focused as I grow in this field. An AI and ML enthusiast, I love working on projects and research, always aiming to deliver meaningful results through data.
SKILLS
Resume
Education
Northeastern University
Sep 2023 – May 2025 • Boston, MA
Master of Science — Data Analytics Engineering
- Coursework: Deep Learning, Neural Networks, NLP, Reinforcement Learning
- Data Analytics, Data Management, Data Mining, Algorithms
- Data Visualization & Computation
PES University
Aug 2019 – May 2023 • Bangalore, India
Bachelor of Technology — Computer Science & Engineering
- Coursework: Statistics, Big Data, Machine Learning, DBMS, Information Retrieval
- Linear Algebra, Data Structures, Operating Systems, Cloud Computing, Software Testing
Professional Experience
Unthink Inc
Jun 2024 – Dec 2024 • Dallas, TX
Data Engineering Intern
- Built ETL pipelines in Python to clean and aggregate high-volume order data from MongoDB and load into AWS S3, reducing dashboard load time by 40%.
- Collaborated with frontend engineers on real-time API integrations, cutting sync delays by 30%.
- Optimized MongoDB and Snowflake query performance with indexing & schema design, improving API response times by 25%.
Tech Mahindra
Jan 2023 – May 2023 • Bangalore, India
Data Analyst Intern
- Designed Tableau dashboards for monthly operational metrics.
- Performed churn analysis with Python models, reducing customer attrition by 4%.
- Cleaned and validated datasets to enhance reporting accuracy.
- Conducted statistical analysis on 2019–2023 data, presenting insights for strategy planning.
Maruthi Technics
Jul 2022 – Dec 2022 • Bangalore, India
Data Analyst Intern
- Analyzed defect data with QC teams, identifying root causes and reducing defects by 12%.
- Forecasted raw material demand with time-series models, reducing inventory costs by 20%.
Projects
Medical Named Entity Recognition using NLP
Extract diseases, symptoms, medications, and treatments from clinical text with transformer models (BERT/BioBERT).

Object Detection using Neural Networks
Real-time detection pipeline with YOLOv5 and Faster R-CNN to identify and classify objects accurately.

Intoxication Detection using Speech
MFCC feature extraction + transformer-based models on 1000+ audio samples. Published at IEEE.

Fake News Detection and Sentiment Analysis
NLP + classification models achieving ~85% accuracy on tweet-level misinformation detection.

Warehouse Management Analytics
Workflows for inventory and operations KPIs; demand and stock movement insights.

Volunteer Matchmaking
Matching volunteers to opportunities using profile features and ranking logic.

Customer Segmentation (RFM)
RFM scoring to create customer cohorts for retention and targeted marketing.

EEG Classification Model
Signal processing + ML pipeline for EEG time-series classification.

Crime Analysis
Exploratory data analysis and visualization of crime trends and hotspots.

Health Conditions Among Children
Notebook analysis of pediatric health indicators and risk factors.

Earthquake Analysis (Python)
Time-series and geospatial exploration of earthquake events.

Face Mask Detection
Computer vision model to detect mask usage from images/video frames.

Blood Bank Management (DB)
Database design + UI for managing donors, inventory, and requests.

Stroke Prediction
Feature engineering + classification models for stroke risk prediction.