Skip to content
Syed Monis Sarwar

Monis Sarwar

Full-stack engineer, building AI agents

I'm a full-stack engineer who enjoys the hard parts: data pipelines, AI agents and the backends that hold them together. Mostly TypeScript and Python, on Postgres.

Most of my projects start from a real problem, not a tutorial. Teachers losing evenings to lesson plans, airfares the official index samples only once a month, ten tabs open just to keep up. I dig into why the problem exists, then build the fix end to end.

My social links if you wish to connect with me

Activity

@MonisMS
GitHub contributions per month, last 12 months
MonthContributions
October 202526
November 202554
December 202537
January 202645
February 202682
March 202672
April 202659
May 202628
June 202633
July 20264
August 202647
September 2026214
October 202693

Projects

Recent work, with the numbers behind it.

All projects
APIx screenshot24,448 fares collected

APIx

Solo build

Collects airfares every day across 12 DGCA-weighted routes and 5 advance-purchase windows, then computes a Jevons/Young price index that reproduces MoSPI's published worked examples to four decimal places.

Python / Next.js / PostgreSQL / GitHub Actions / d3-geo

Curio screenshot275+ sources · 26 topics

Curio

Solo · 71 commits

Topic-based feeds pulled from 275+ sources across 26 topics. Five scheduled jobs run ingestion, email digests, source-quality scoring, cleanup and re-engagement, and one slow feed never stalls a run.

Next.js / TypeScript / Drizzle / PostgreSQL / Resend / GitHub Actions

FieldProof screenshot40 tests on a real database

FieldProof

Code Cubicle 6.0 · team of 4

A Python worker tags uploaded photos with CLIP, embeds them for pgvector search, assigns sites from EXIF GPS and estimates before/after green cover. AI-written reports that invent a number are rejected.

FastAPI / Python / PostgreSQL / pgvector / CLIP / LiteLLM / Next.js

DataPilot screenshot9 sources in parallel

An LLM planning agent turns a prompt into a collection plan, runs robots.txt-aware collectors across 9 sources in parallel, then validates, de-duplicates and confidence-scores the rows into a Postgres dataset.

Next.js / TypeScript / PostgreSQL / Cheerio / LLM agents

Shikshak Saathi screenshotEnglish + Hindi

Shikshak Saathi

Hack-e-Awadh 2026 · team of 3

Turns an NCERT chapter into objectives, a lesson plan, a worksheet and a quiz tagged by misconception, with a validator that auto-repairs malformed AI output, Word export and Hindi voice.

Next.js / Prisma / PostgreSQL / Better Auth / Gemini / Sarvam AI

Open source

Merged pull requests to projects I don't own.

Smaller builds

Focused tools and systems worth a look.

Contact

Have a role, a project or a problem worth solving? Send a note, I read every one.

Book a call
Open to work