Skip to content

VYNN AI

A personal financial analyst for every investor

founder & sole engineer · Jul–Dec 2025 · 5K+ registered users

See it work

Real runs on 15 October 2026, with the same words sent to both. Sped up; the clocks show real elapsed time, and Claude’s wait for permission is not counted. The full comparison

Why it exists

I invest my own money, and with a full-time engineering job I never had time to read all the news that moves it. I wanted an analyst that watches the market for me around the clock, works out what a critical event means for what I own, and tells me before the price has already moved. Research like that is sold by the seat, and the AI tools built to replace it make up their numbers.

What I built

So I built every investor a personal financial analyst. Name a company, a fund, a coin or a prediction market, in any language. VYNN reads the filings and the news, builds the valuation model and writes the report in about two minutes, for about three cents. Code computes every number, and every claim carries its source. I built every part of it and run it myself.

Where it’s going

Next, the part I started it for: VYNN stops waiting to be asked. It will watch every portfolio around the clock and tell investors the moment news, earnings or a price move changes the case for what they own.

Impact

  • 5K+ investors signed up, from a 500-person pilot.
  • Six to twelve hours of an analyst’s research in about two minutes, for about three cents.
  • Every release is checked against a set of reference valuations before it ships.
NVIDIA’s page in the VYNN app on 30 September 2026: the price and a one-month chart on the left; on the right the thesis of record (version 9, fair value $356.54, Strong Buy at low confidence, what the price requires) and the Street’s consensus in a separate card.

NVIDIA’s page in the app on 30 September 2026, whole. Choose a number, here or on the page, to look closer.

VYNN, as an investor sees it. NVIDIA’s thesis of record of 30 September 2026, captured from the app. A re-run adds a new record: NVIDIA’s latest is on vynnai.com.

More real output

VYNN AI · complete system architecture

Scroll across to follow the full system

The company website leads to the React product. Caddy routes requests to FastAPI, which authenticates jobs and launches research workers. A tool-use agent selects research and valuation tools; the report is validated before output. MongoDB, Redis, and persistent files support the API and workers.FastAPI · identity, job lifecycle, streams, and research recordsResearch worker · ephemeral container, pinned image, user-scoped volumeDocker SDK · queued and bounded execution · progress over SSECompany websitevynnai.comNext.js · product, mission, demoResearch workspace · app.vynnai.comReact / TypeScript · Vite / Nginx · analyst chat + market terminalCompanies · funds · digital assets · prediction markets · positionsCaddy · HTTPS and reverse proxyREST · server-sent events · price and news WebSocketsIdentity & accessOAuth · ownership checksRedis-backed quotasResearch jobsQueue · launch · cancelReconcile after restartMarket streamsShared price/news pollingCache-backed universeSaved researchImmutable thesis recordsRevisioned portfoliosTool-use agent · research specialists & valuation engineReads the question, selects the work, and calls tools with structured argumentsFinancial dataFundamentals · filings · unitsShared typed FinancialStateSymbolic valuationSourced cost of capitalDCF formulas and assumptionsNews intelligenceRetrieve · rank · screenRuns alongside model workReport generationSix sections composed in parallelCalculator → evidence → narrativePublication checksClaim validation and citation coverageFlag a value the analysts do not backLive DCF workbookTen Excel sheetsEditable inputs and formulasSourced reportReport · PDFValuation, catalysts, and risksStructured resultsReturned to the APIStored as a dated thesisMongoDBMarket profiles, articles, conversationsImmutable theses and portfoliosShared article DAO in vynn-coreRedisQuotas and expiring job snapshotsCross-instance read-throughOnly the owner republishes job statePersistent filesResearch outputs and model inputsPer-user worker mountsAtomic reference-data cache writesStorage serves the API and research workers; records and cache have separate lifecycles.Hetzner · Docker · persistent volumesDigest-pinned worker images · bounded concurrency · owned artifact recovery
Explore the individual layers

Explore the stack

Research, calculate, then check the claim

A tool-use agent selects financial, market-data, and analysis tools, with specialists for deeper research. Financial statements feed a symbolic valuation engine; model and news work can run in parallel before the report is assembled.

Calculation
Code computes valuation and recommendation fields. Sourced market inputs ground the cost of capital; the exported ten-sheet workbook retains its formulas.
Publication
Evidence and citation checks gate the narrative. A value the analysts do not back is still published, at low confidence, with an alert that states both positions.
Research and valuation engine

Decisions behind the product

The model never writes a number

A deterministic DCF engine computes the valuation, and a formula evaluator reads the workbook back. The model writes the explanation. A validator rejects any figure that doesn’t match the calculator, and any write-up that cites fewer than 95% of its claims.

It stands by its value

When well-covered analysts disagree, VYNN still publishes its fair value and rating, at low confidence. An alert shows both numbers on the report’s first page, in the workbook and in the chat.

Discount rates from real market data

The cost of capital is built from published rates in the cash flows’ own currency (about 20 currencies), with Damodaran’s risk premiums and a beta measured against the company’s home index. A Tokyo listing is valued in yen, against Japanese rates.

Four repositories, one system

stock-analyst

Research engine · Python

The research engine behind every report.

  • One agent chooses its tools for each question
  • Four specialist agents research in parallel
  • A deterministic engine computes every number

Python · LangGraph · Tool-use agents · Symbolic DCF · MongoDB

Agentic-Analyst/stock-analyst·DeepWiki

api-runner

Control plane · FastAPI

The control plane: each analysis runs in its own locked-down container.

  • Jobs survive restarts
  • Progress streams live, without extra load
  • Every record belongs to one user

Python · FastAPI · Docker · MongoDB · SSE · WebSocket · OAuth · Redis

Private repository

vynnai-web

Product interface · React + TypeScript

The app: research chat beside a dashboard for companies, funds, crypto and prediction markets.

  • Every number keeps its source, date and currency on screen
  • A gap with the analysts stays next to the number
  • Missing data is shown as missing

React · TypeScript · Vite · TanStack Query · WebSocket · Recharts

Private repository

vynn-core

Shared persistence · Python

Shared storage for the API and the research workers.

  • One identity per article, from its canonical URL
  • Every timestamp in UTC
  • One implementation, so the two services can’t disagree

Python · MongoDB · Redis · Canonical URLs · UTC / BSON

Agentic-Analyst/vynn-core