CryptoTalkies
AI-powered crypto news, education and market-intelligence platform — web and mobile
A free crypto intelligence site and Flutter app: live market data, AI signals, sentiment and whale tracking, a multi-provider AI agent, and automated news and education content.
Status: Website live and free to use; Android app published on Google Play. · Period: 2025 – present · Source: Private repositories (web and mobile).
01Overview
CryptoTalkies connects markets, signals, sentiment, whale activity and AI analysis on one platform. The web app is a statically exported Next.js site backed by Firebase Cloud Functions. The Flutter app shares the same Firebase project and AI functions.
02The problem
Retail crypto learners have to piece together prices, news, sentiment and technical analysis from many sources, often in jargon. CryptoTalkies brings them together with plain-language AI explanations and educational content.
03Who it is for
- Crypto learners and enthusiasts
- Mobile users following markets on Android
- Readers of crypto news and guides
04My role
Sole builder of the web platform, the Flutter app, the Firebase back end, the AI orchestration and the content automation pipelines.
05Implemented features
- Live prices, rankings and coin sentiment, plus a Fear & Greed index.
- AI trading signals from multi-timeframe technical indicators (RSI, MACD, Bollinger Bands).
- Whale tracker, AI coin screener, narrative detector, arbitrage scanner and portfolio optimiser.
- ARIA, an AI crypto agent that pulls live prices, sentiment and headlines, with fallback across AI providers.
- News tagged by AI for likely market impact, a daily AI market brief, guides, a glossary and quizzes.
- Flutter app with portfolio tracking, price alerts, push notifications and 12 languages.
06System architecture
A statically exported Next.js site and a Flutter app share one Firebase back end. Cloud Functions orchestrate market-data APIs and several AI providers.
07Engineering challenges and approaches
Reliable AI answers across providers
ARIA orchestrates several model providers with a fallback chain, so a provider outage or rate limit degrades the answer instead of failing the request.
One back end, two clients
Web and mobile share one Firebase project. Functions are grouped into web and mobile codebases with explicit naming, so each client can be deployed independently while sharing the AI and data functions.
Content at scale without a newsroom
Scheduled pipelines ingest news, extract facts, deduplicate and generate articles. Separate pipelines handle SEO monitoring, indexing submissions and short-video generation.
08Outcomes and current status
- Website live and free to use, with no account required.
- Android app published on Google Play.
- Large programmatic SEO surface (coin, exchange and guide pages) generated at build time.
No usage or revenue numbers are published here unless I can verify them.
09Roadmap (not yet shipped)
- iOS release of the Flutter app.
- Further personalisation of alerts and briefs.