Hi, my name is

I'm 

AI Solutions Engineer in New York City & Carnegie Mellon Statistics + Machine Learning grad. I build AI automation systems, data pipelines, and full-stack products that streamline overly complicated procedures into actions.

01. About Me

I'm a builder at the intersection of machine learning and software engineering. At Kokka & Company I re-architect supply-chain data flows with multi-agent orchestration, wire up SAP ERP and Salesforce CRM into inventory-aware demand forecasters, and ship analytics tools non-technical teams actually use.

Before that, at Nagarro, I built full-stack "Living Persona" simulation platforms and NLP pipelines for pharmacovigilance auditing — from ChromaDB semantic search to automated stakeholder reporting.

I studied Statistics & Machine Learning at Carnegie Mellon, and I speak English, Hindi, and Mandarin. When data gets messy, that's when it gets fun.

0%
accuracy — neural net built from scratch
0k+
training samples classified in < 1 min
0+
programming languages
0
spoken languages

02. Where I've Worked

AI Solutions Engineer

Kokka & Company · Pittsburgh, PA

Feb 2026 — Present

SMB Consulting

  • Re-architected multi-channel supply chain data flows with multi-agent orchestration for automated logistics communication.
  • Dynamically pulled data from SAP ERP and Salesforce CRM for an inventory-aware demand forecaster.
  • Deployed local OpenClaw instances for SMS-based customer service — a working analytics-backed tool for a non-technical audience, no help-desk overhead.
Multi-agentSAP ERPSalesforceForecasting

Developer

Nagarro · New York City, NY

Jan 2025 — Jan 2026

Living Personas — Simulated B2B & B2C Engagements

  • Built a full-stack app with TypeScript, React + Tailwind (Vite) and Node.js + Express / Flask services, containerized with Docker.
  • Designed data pipelines on GCP Cloud Run turning social & engagement data into analytics-ready persona profiles.
  • Implemented authentication, secret management, and cloud storage for secure persona data and chat history.

Pharmacovigilance Audit & Campaign Management Bot

  • Built ingestion + retrieval with ChromaDB semantic search over large unstructured datasets, using transformers, spaCy, and NLTK.
  • Designed a query processor for classification & entity extraction, integrating the OpenAI API to auto-summarize findings for stakeholders.
ReactPythonGCPDockerChromaDBNLP

B.S. Statistics & Machine Learning

Carnegie Mellon University · Pittsburgh, PA

Aug 2020 — Jun 2024
  • Coursework: NLP, Algorithms & Advanced Data Structures, AI: Representation & Problem Solving, Machine Learning, Statistical Methods in Finance, Human-Centered Software.
Machine LearningNLPStatistics

03. Things I've Built

★ —

MNIST Digit Recognizer

Neural-network digit OCR written from scratch in Python — no ML frameworks. Hits 85% accuracy across 60k+ samples and trains in under a minute.

PythonNumPyNeural Nets
★ —

OpenAI Gym RL Experiments

Reinforcement-learning experiments in OpenAI Gym — training agents to solve control environments, explored in Jupyter notebooks.

JupyterRLGym
Report

Video Game Market Analysis

Cleaned and analyzed large market datasets from VGChartz & NPD, applying MDS and Pearson's tests — visualized trends with mosaic & ridge plots for business-ready insights.

RStatisticsData Viz
Explore all repos on GitHub →

04. My Toolkit

Python JavaScript C / C++ Java R Ruby SQL PyTorch scikit-learn Transformers spaCy NLTK Pandas NumPy PySpark ChromaDB React Node.js Flask Vite Tailwind Docker Streamlit GCP Azure Power BI Excel

05. Resume

Want the full background in one place? You can open my resume in the browser or download a copy.

View resume Download resume

06. Get In Touch

I'm always up for talking about data and AI problems, or a role where I can build things that matter. My inbox is open — say hi!

Open mail app
Pittsburgh, PA & New York, NY