VoooAI – natural language to multi-modal AI pipelines
Pseudocode IR layer validates before execution, but orchestrates existing APIs like Midjourney and Runway.

Natural language to trading logic beats manual indicator configuration.
Retail traders, crypto traders
TradingView · TrendSpider · Coinrule
I've been building CandlePulse, a platform that lets traders create trading alerts using natural language instead of manually configuring indicators.
For example, a trader can write something like: “Alert me when BTC forms an indecision candle near support with high volume.” The system converts that description into an alert logic that runs against market data.
Some of the things the beta supports right now: • Natural language alerts – describe a trading condition and the system converts it into an alert rule. • Chart-based alerts – draw support/resistance lines on the chart and create alerts based on price interaction with those levels. • Market scanners – scan multiple symbols for a given alert condition. • AI chat assistant – ask questions about the chart or market data. • Email alerts – notifications are sent when the conditions are triggered.
The goal is to make it easier for traders to express ideas without needing to code strategies or manually configure complex indicators.
I’m currently opening up the beta and would love feedback from traders and developers here.
You can try the beta here: https://candlepulse.com
Happy to answer questions about the design, architecture, or roadmap.
Pseudocode IR layer validates before execution, but orchestrates existing APIs like Midjourney and Runway.
Replaces every function with English prompts executed by Claude, tests still pass.
Yet another AI trading wrapper in a space crowded by funded incumbents.
Author calls it 'super shitty V0 MVP'—embryonic but the prediction-to-trade angle is real.
Natural language E2E tests sound good until you need debugging or maintenance.
5 ladder orders in 1 sentence instead of 50 clicks — 10 users, 60K orders shipped.