I build systems that hold up outside the demo.

I'm Aditya Nair, a Computer Engineering student at the University of Mumbai. My work sits between backend systems, data pipelines, and applied AI.

Mumbai, India
01

Now

Research

Currently, no ongoing research.

SimPPL

Working as a Research Engineer Intern architecting event-driven data pipelines and collectors for Arbiter (backed by UN, MIT, & Stanford).SimPPL

02

Experience

Sept 2025 - Present

Research Engineer Intern

SimPPLMassachusetts, USA
  • Worked on Arbiter, a social intelligence platform backed by the United Nations, MIT, and Stanford.
  • Architected serverless, event-driven data pipelines plus staging and testing infrastructure for prediction market analysis.
  • Built asynchronous collectors for X, Bluesky, Reddit, YouTube, and Truth Social, supporting analysis of 100M+ posts.
Distributed Systems
AWS
Event-Driven
Python
Data Pipelines
Jul 2025 - Aug 2025

Open Source Research Collaborator

SimPPLUSA (Remote)
  • Researched and analyzed data collection approaches for Gab, Bluesky, and Truth Social.
  • Created a Truth Social data collector for scalable post collection.
  • Deployed the collector on AWS to support reliable, scalable ingestion.
Open Source
Data Collection
AWS
Python
Social Graphs
Jul 2025 - Aug 2025

Inaugural Veritas Research Fellow

The Midas ProjectUSA
  • Aggregated fragmented OpenFEC and state-level campaign finance data to track over $100M in AI industry political spending.
  • Engineered an NLP regulatory capture engine using text embeddings to quantify legislative alignment.
  • Used NetworkX graph analysis to uncover overlapping funding clusters between AI and crypto sectors.
NLP
Embeddings
NetworkX
Campaign Finance
Python
May 2025 - Jul 2025

SDE Intern

Mass International LogisticsDammam, Saudi Arabia
  • Designed a full-stack logistics management platform across system design, data modeling, and API architecture.
  • Built Next.js and React interfaces with Node.js and Express APIs backed by PostgreSQL on EC2 and DynamoDB.
  • Implemented Redis caching and AWS Lambda functions for customs document parsing.
Next.js
React
PostgreSQL
DynamoDB
Redis
AWS Lambda
03

Projects

LexAI
LexAI preview
ArchitectureDetails

LexAI

Fine-tuned Llama 3.2-3B-Instruct on curated legal corpora using LoRA and QLoRA with 4-bit quantization. Built a legal document intelligence SaaS with PDF annotation overlays, clause-level risk extraction, and serverless GPU inference.

143
Next.js 14
React 19
TypeScript
Convex
+3
Cortexa
Cortexa preview
ArchitectureDetails

Cortexa

Chat with your class notes on WhatsApp to instantly get summaries, answers, and PDF study guides from your own material.

184
Python
Twilio API
OpenAI API
LangChain
+2
MeTube
MeTube preview
ArchitectureDetails

MeTube

Not a clone but rebuilt the entire YouTube, implementing responsive video playback, user channels, video uploads with transcoding, comments, likes, and search.

225
React
TypeScript
Node.js
Express
+3
Ivigil-Ease
Ivigil-Ease preview
ArchitectureDetails

Ivigil-Ease

Real-time online proctoring system using OpenCV, MediaPipe, and Matplotlib to track eye movements, detect suspicious activity, and generate session reports with 96.25% tested accuracy.

318
React
Python
OpenCV
MediaPipe
+3
Warmpath
Warmpath
Details

Warmpath

An active TypeScript systems project exploring low-latency data access patterns, structured memory caching, and efficient server-client communications.

92
TypeScript
Node.js
Systems Design
Caching
+2
Data Donation
Data Donation preview
ArchitectureDetails

Data Donation

A browser extension and backend that anonymously collects and analyzes user engagement data from Instagram and X (Twitter) to support social media research with privacy and gamification.

123
JavaScript
Extension APIs
Node.js
Express
+2
04

Education

University of Mumbai

Bachelor of Engineering in Computer Engineering

Aug 2023 - May 2027
05 — Contact

Have a role, project, or research problem worth building properly?