AI Trainer, Data
Micro1 · New York, NY · Sep 2026 to Present
detect and transform pii in real-world datasets (json, csv, slack exports, postgres records) to produce clean, privacy-safe training data for ai models.
MS Computer Engineering · NYU Tandon
Software Engineer building ML pipelines, backend systems, and the full stack products that sit on top of them.
Micro1 · New York, NY · Sep 2026 to Present
detect and transform pii in real-world datasets (json, csv, slack exports, postgres records) to produce clean, privacy-safe training data for ai models.
Tiny Archives · New York, NY · Sep 2026 to Present
Developing full-stack features for ArchiveOS, a web-based archival collection management platform. Architecting React/Next.js frontend components and optimizing Django backend; assuming scoped product ownership for assigned features including collection inventory, global search, admin dashboards, and Stripe integration.
BigC Technologies · Bengaluru, India · Jul 2023 to Jun 2024
Built and shipped ML pipelines end to end: data ingestion, model training, and deployment on cloud infrastructure, with a focus on scalability and cutting training time on production workloads.
BigC Technologies · Bengaluru, India · Mar 2023 to Jun 2023
Developed and deployed supervised learning models, including logistic regression and random forest, to detect patterns across large scale datasets.
Wayfinding tool surfacing real ADA accessibility status for all 496 NYC subway stations from MTA's open dataset, automatically routing to the nearest accessible alternative using a Haversine ranking algorithm.
Interactive dashboard for automated ML workflows: upload a dataset, run model selection and hyperparameter tuning, and visualize performance, all without writing code.
Human in the loop moderation platform combining NLP classification, confidence based routing, a moderator dashboard, audit logging, and bias monitoring, designed as a full socio technical system.
Full stack developer tools marketplace where developers browse, submit, and discover tools by category and stack, with a REST API backend and a responsive frontend, built end to end.
Classification pipeline predicting student stress levels from 5,000+ behavioral and academic records, benchmarking four models on F1 and ROC AUC, hitting 84% accuracy.
End to end ML pipeline for retail sales forecasting on 8,000+ records, an 18% accuracy improvement over baseline through feature engineering and tuning.
Distributed pipeline streaming 2,000+ real time sensor events per day via ESP32 and MQTT, with sub 200ms end to end latency into AWS.
NLP based fraud detection classifying fake social media accounts at 84% accuracy across 10,000+ records, built as part of a KSCST funded research initiative.
Python / JavaScript / SQL / C++ / HTML & CSS
TensorFlow / Scikit learn / NLP / Pandas / NumPy / Model Evaluation
FastAPI / Node.js / REST APIs / Flask
Data Pipelines / Feature Engineering / Streamlit / SQL
AWS / IBM Cloud / MQTT / Git & GitHub
MS, Computer Engineering · NYU Tandon · Sep 2024 to May 2026
Coursework spanning systems engineering, machine learning, full stack development, and human centered design.
B.Tech, Information Science Engineering · Jun 2019 to Aug 2023
Coursework spanning data structures, algorithms, databases, and software engineering. Led a KSCST funded research initiative.
Built and shipped a project under time pressure alongside engineers from across the industry.
IBM Academic Initiative certification, issued 31 Oct 2022. Verify on Credly ↗