Fintech and payments Case study

Reactor: AI assistant and Web3 data back end

A Node.js back end for Reactor, an AI chat assistant with Web3 tools: streaming chat and voice, a keyed language-model API for businesses, and a crypto service for token audits, swaps and rewards.

Reactor: AI assistant and Web3 data back end: website
Reactor chat design on a phone: a prompt, a streamed answer with response statistics and suggestion chips
Client
Reactor
Service
Web apps, Integrations, AI
Years
2024–2025
Stack
OpenAI, TypeScript, Node.js, PostgreSQL and 14 more
Platforms
Web

Overview

Reactor is an AI assistant that combines everyday chat with tools for crypto users. We built its back end in Node.js: user accounts and single sign-on, real-time streaming chat with stored conversations, file uploads and streamed text-to-speech, a separate keyed API that businesses use to reach the same models, and a crypto service for market data, token security checks, swaps and rewards.

The challenge

Answers have to stream in real time over WebSockets, the same models are offered to consumers and, with limits and usage accounting, to business customers, and the crypto side needs fresh market data and token checks from several outside sources.

What we built

  1. Accounts, several sign-in methods and single sign-on for client apps
  2. Real-time chat over WebSockets with streamed answers and stored conversations
  3. File uploads and real-time text-to-speech streaming
  4. A business API with API keys, rate limits, token accounting and a usage analytics dashboard
  5. Language-model routes across several providers, and a retrieval prototype with LangChain and Ollama
  6. A crypto service: market data, token security audits, swaps and limit orders, portfolio history and discovery
  7. Quests and reward tracking with smart-contract interaction, and reward reports with Excel export
  8. Notifications by Firebase, Expo and chat platforms, and scheduled jobs
  9. Subscription and payment flows with Stripe card payments and auto-charge, webhooks for iOS and Android purchases, and plan emails for purchases and cancellations
  10. Swagger documentation for the business API

How a system like this fits together

Reference architecture: AI assistant and Web3 data back end

Simplified reference architecture, not a client schematic.

Reference architecture: AI assistant and Web3 data back endZones from left to right: clients (chat, business and crypto clients), the application (user and chat API, business API, crypto service), data (PostgreSQL, MongoDB, Qdrant) and third-party services (language models, market and security data, blockchain network). Full description in the numbered list below the diagram.ClientsApplicationDataThird-party servicesWebSocketsign-inAPI keyrequestsaccounts, chatsmodel callskeys, usagemodel callscrypto datamarket data, swaps, rewardsvector searchChat clientWeb or mobilechat appBusiness clientCalls withan API keyCrypto clientWeb or mobileappUser and chat APIFastify: accounts, SSO,chats over WebSocket,uploads, speechBusiness APIExpress: API keys, ratelimits, usage analyticsCrypto serviceFastify: market data,token checks, swaps,portfolio, quests,rewardsPostgreSQLAccounts, chats,API keys, usageMongoDBCrypto and rewardsdataQdrantVector search forthe AI assistantLanguage modelsOpenAI, Vertex AI,OllamaMarket and security dataPrices, token checks,swap quotesBlockchain networkSwaps and rewardcontracts via ethersReference architecture: AI assistant and Web3 data back endZones from left to right: clients (chat, business and crypto clients), the application (user and chat API, business API, crypto service), data (PostgreSQL, MongoDB, Qdrant) and third-party services (language models, market and security data, blockchain network). Full description in the numbered list below the diagram.Chat and business APIsCrypto servicesign-in, WebSocketaccounts, chatsmodel callsmodel callskeys, usageAPI keyrequestsswaps, rewardsmarket datacrypto datavector searchChat clientWeb or mobile appUser and chat APIFastify: accounts, SSO,chats over WebSocket,uploads, speechPostgreSQLAccounts, chats, keys, usageLanguage modelsOpenAI, Vertex AI, OllamaBusiness APIExpress: API keys, ratelimits, usage analyticsBusiness clientCalls with an API keyCrypto clientWeb or mobile appMarket and security dataPrices, token checks, quotesBlockchain networkSwap and reward contractsCrypto serviceFastify: market data,token checks, swaps,portfolio, quests,rewardsMongoDBCrypto and rewards dataQdrantVector search, AI assistant
One chat API streams answers from language models and stores conversations; a keyed business API serves the same models with limits and usage analytics; a separate crypto service handles market data, token checks, swaps and rewards.
  1. A user signs in to the chat app (password, social login or single sign-on) and opens a WebSocket to the user and chat API, which also takes file uploads.
  2. The API checks the account and loads the stored conversation from PostgreSQL.
  3. The prompt goes to a language-model provider. The answer is streamed back over the WebSocket, saved to the conversation and, for voice, streamed on as text-to-speech audio.
  4. A business customer calls the keyed business API instead: the API key is checked, the request is rate-limited, the model call is made and usage is recorded in PostgreSQL for analytics.
  5. Crypto requests go to the crypto service, which pulls market data and token security checks from outside sources, calls smart contracts for swaps and rewards, and keeps crypto, quest and reward data in MongoDB (vector search in Qdrant).

Technology used

Technology named in the project records, by layer
LayerChoicePurpose
AIOpenAI, Vertex AI, Ollama, LangChainchat, completions and retrieval
APIsNode.js, Fastify, Express, TypeScriptuser, chat, business and crypto services
DataPostgreSQL, MongoDB, Redis, Qdrantaccounts and chats, crypto data, cache, vector search
PaymentsStripecard subscriptions and auto-charge
DeliveryDockerbuild and deploy
Real timeWebSocketsstreamed chat answers and speech
Web3ethersswaps, signing and rewards

What happened next

  • Delivered

    The work described above was delivered to the client.

  • Operational facts
    • Our work on this project spanned 2024–2025.

Could this be your project?

  • You need a web application with user accounts, admin tools and reporting. Website and web app development
  • Your product has to connect to payment, accounting, messaging or other third-party systems.
  • You work in fintech and payments and want a team that has built for it before. Fintech and payments work

Request similar work

Questions about this project

How long did this project take?

Our work on it ran from 2024 to 2025.

What technology was used?

OpenAI, TypeScript, Node.js, PostgreSQL, MongoDB, Redis, Stripe, Docker, Fastify, Express, Qdrant, LangChain, Ollama, Vertex AI, ethers, Socket.io, Sequelize, Swagger.

Similar work

Also TypeScript and PostgreSQL, same industry

  • Fintech and payments
  • Web apps
  • Integrations

2024–2025

A .NET trading engine, API and Next.js control console for Rainbow Capital that runs configurable US equity options strategies under per-underlyer risk limits, with daily profit and loss reports to Slack.

Banner with the Xpress Money logo and the title Global remittance platform, over a globe with transfer routes and Transfer sent and Money received notices

Also TypeScript and Redis, same industry

Xpress Money: global remittance platform

  • Fintech and payments
  • Business software and compliance
  • Web apps
  • Integrations

2022–2023

Xpress Money, a remittance platform for Wizz Financial: Angular portals and .NET services deployed on Azure, with the commercial ESB replaced by Azure Service Bus.

RoofReveal home screen design with three report tiers: Canvassing, Explorer and Xactimate, each over an aerial photo of roofs

Also Node.js and Stripe, also a web app

RoofReveal: roof measurement and reporting platform

  • Business software and compliance
  • Web apps
  • AI
  • Integrations

2024

A web app that finds a roof from an address, shows detected roofs on a satellite map and sells tiered reports, with a Python service using Google Solar API data and Gemini roof-photo analysis.

Start a similar project

Tell us what you need. We come back with questions, what is realistic, an estimate and the closest work we have done.