№ 00 — Adam Manji · Berkeley, CA

Founder& engineer.

Adam Manji

Building in fintech and sports — a B2B AI underwriting platform (Bashi) and a real-money daily-fantasy app (Snappy) currently in pre-seed.

BasedBerkeley, CA
EducationUC Berkeley · Data Science
Shipping2 startups, pre-seed
Emailadam@bashi.app
The work, indexed
A short biography

I studied data science at UC Berkeley, and I now spend my days writing the code, the pitch decks, and the cold emails for two pre-seed companies that have nothing in common on the surface — and almost everything in common underneath.

i.

Fintech

Bashi turns a multi-week credit-underwriting workflow into an afternoon. Claude-powered extraction, deterministic financial modeling, and an output that looks like the memo a senior credit officer would write.

ii.

Sports

Snappy is built on a contrarian bet — that the right form-factor for daily fantasy is short, head-to-head, and feels like a mobile game, not a spreadsheet. Live matchmaking, snake drafts, real-time scoring.

iii.

The seam

Both products live where messy real-world data meets product surface. That's the work I'm interested in: turning the hard, schema-less middle of a problem into something a normal person can use.

The Work, Indexed

№ 01A case study

Bashi

An AI copilot for commercial credit underwriting.

bashi.appadam@bashi.app
bashi/extract.run
live
→
Bashi product screenshot 1
fig 01
Bashi product screenshot 2
fig 02
Bashi product screenshot 3
fig 03
§ 01

The problem

Commercial credit underwriting is a multi-week ritual. Analysts manually re-key numbers from PDF financials into Excel, build 3-statement projections, and write a memo. The work is tedious, error-prone, and low-leverage — the slowest part of every middle-market deal.

§ 02

The wedge

Bashi reads the PDF the way a credit officer would. It picks the right pages with vision-language models, extracts 51+ metrics with type-checked extractors, builds the deterministic 5-year projection in pure Python, and writes a memo that you'd be willing to put in front of a credit committee.

§ 03

Why it matters

We turn a two-week underwriting pass into an afternoon. The model does what models are good at — read mess, structure it. The deterministic layer does what models shouldn't — arithmetic, projections, memo skeleton. The human does the only part that still matters: judgment.

Stack9 items
  • FastAPI
  • Python 3.11
  • Claude Sonnet 4.6
  • pdfplumber
  • openpyxl
  • ReportLab
  • PostgreSQL
  • Railway
  • pytest (716)
№ 02A case study
A daily-fantasy reset

Snappy.

Real money. One quarter. Two players. Done.

snappyfantasy.comadam@snappyfantasy.com
snappy/Q3.live
5:42 left
you28
opp25
  • S. Curry12
  • L. Dončić9
  • A. Edwards7
  • J. Tatum11
  • S. Gilgeous-A.8
  • G. Antetokounmpo6

fig 01 — Snappy live-scoring engine, simulated

Snappy Home screen
Home
Snappy Draft screen
Draft
Snappy Live screen
Live
§ 01

The thesis

DraftKings won by being a spreadsheet. The next generation of fantasy users won't sit down for a 12-hour slate. They'll play in the queue at Chipotle. The product needs to feel like a mobile game, finish in minutes, and make every play feel consequential.

§ 02

The form factor

Snappy ships two formats: NBA quarter-only contests that resolve in roughly 30 minutes, and MLB head-to-head snake drafts that finish in three innings. Both are 1v1 by default. Both have a 20-second draft turn timer. Both feel — by design — like a video game.

§ 03

The under-the-hood

Live matchmaking with bot fallback. Double-entry ledger across cash and bonus wallets. Glicko-2 ratings that bucket players into divisions from Bronze to Legend. Three independent scoring engines. Real-time stat polling against league feeds. 563 backend tests so the money side never slips.

Stack9 items
  • React Native / Expo
  • FastAPI
  • PostgreSQL
  • Auth0
  • Glicko-2
  • WebSockets
  • BallDontLie API
  • Railway
  • EAS Build · TestFlight
№ 03Data analysis

A 38-point shift, in sixteen years.

Six elections, two data sources (Census CPS and Edison exit polls), one stubborn pattern. The Hispanic male presidential margin has collapsed from D+31 in 2008 to R+12 in 2024 — a 43-point realignment, with 35 of those points coming in the last cycle alone.

6
elections
38pt
net shift
2
data sources
Hispanic male margin (D — R)y axis: pts
+30+20+100-102004D+92008D+312012D+242016D+192020D+92024R+12

fig 01 — Method: simple two-party margin from CPS-weighted exit poll crosstabs. Recreated for this site; full code & data on request.

Etc.— archived & assorted

  • 042025

    Fantasy-Duel

    Backend · API design

    FastAPI + Postgres matchmaking service for 1v1 fantasy contests. Built the data model, auth, and migrations end-to-end before pivoting the idea into Snappy.

    Archived
  • 052022

    TaserTag

    Hardware · IoT

    Arduino-based competitive tag system. Embedded firmware, Python backend for match state, and a thin web frontend for scoring.

    Archived
  • 062022—2023

    CalHacks 9 & 10

    Hackathon · 36 hours

    Two consecutive years competing at UC Berkeley's flagship hackathon. Sleep-deprived shipping, demo-day pressure, public repo.

    Submittedrepo
  • 072023

    Cal-Dining

    Side project · scraping

    Berkeley dining-hall menu aggregation. Quick public utility built for fellow students.

    Archivedrepo
§ V — colophonContact

Get in touch.

Best ways to reach me: investor email at adam@bashi.app for fintech, adam@snappyfantasy.com for sports, anything else at the Berkeley address.