Orderflow Terminal

A personal terminal to study the options market and company financials in one place: flow and gamma exposure from the full chain, price structure, fundamentals read straight from the SEC and an AI summary whose every number is checked against the data.

Personal project2026

Code on GitHub

Period: Sep 2026
Team: Personal project
Context: Personal use, locally and on my server, inside a private network
My role: The whole project
Tech stack: TypeScript
Next.js
React
Fastify
uWebSockets.js
Zod
TanStack Query
lightweight-charts
Tailwind CSS
Vitest
Playwright
Docker

The Flow tab with real data: the CBOE options chain, delayed about 15 minutes, and Yahoo Finance prices.

What it does

Flow: the stock chart with gamma levels (call wall, put wall and zero gamma), supply and demand zones, structure breaks and Fibonacci, next to the options chain and the net call and put premium.
Gamma exposure by strike and expiry, with a heatmap that compares each level with the day’s first snapshot.
Quarterly fundamentals from the SEC, a bear, base and bull valuation from a DCF and the company’s own historical multiples, and an earnings simulator.
On-demand AI analysis that sums up gamma, structure, technicals and consensus, and checks that every number in the text exists in the app’s data.
Market overview: sector map, top movers, market regime and SEC catalysts.

How it is built

A TypeScript pnpm monorepo: a Fastify and uWebSockets.js gateway and a Next.js 15 and React 19 web app that share types and Zod schemas.
Provider-agnostic ingestion: free CBOE data by default, or Interactive Brokers and Schwab for real time, watching for connections that stop receiving data and reconnecting automatically.
Under load, quotes are dropped before trades, and charts render on canvas at 60 fps by batching ticks per frame.
No made-up data: if a source fails, the UI says so; a missing value is shown as a dash, never as zero, and every figure keeps its provenance.
Deployed with Docker on my server and reachable only from my private Tailscale network.

In numbers

automated tests: 356
Playwright UI journeys: 10
spent on data: 0 €

Gallery

AI analysis of NVDA with situation, key levels, scenarios and risks, next to gamma exposure by strike — On-demand AI analysis. At the bottom, the check that every number in the text exists in the app’s data.

NVIDIA quarterly table with revenue, margins, EBITDA and cash flow, next to the earnings simulator — Quarters from the SEC. Q4 is the full year minus Q1 to Q3, because the SEC does not publish it separately.

Sector map with the eleven sectors, top gainers and losers and market regime — Market overview: sectors, the day’s movers and market regime.

NVDA valuation in bear, base and bull scenarios, with the analyst consensus — Valuation in three scenarios from a DCF and the company’s own historical multiples, next to the analyst consensus.

On the phone

NVIDIA quarterly table on the phone — Fundamentals with horizontal scrolling.

Technical signal, analyst consensus and volatility ranges on the phone — Signals and implied volatility ranges.

Let’s talk

I’m from Seville and now live in Madrid. I like things to work end to end: designing the system, writing the code, deploying it and measuring whether it really holds up.

If you’re looking for someone for backend, data or applied AI work, write to me and I’ll get back to you.