My co-founder and I built Tickerterm – making algo trading accessible through AI.

We met at Bubble (employees #2 and #4). Watched AI coding change who could build software. Started wondering: could the same thing happen for algo trading?

The platform: Chat with AI to write your strategy → code deploys serverless → trades execute in our built-in brokerage (Alpaca partnership). 60 seconds from idea to live.

Fully licensed (FINRA/SIPC compliance, NASDAQ/NYSE data licenses). Launched today.

Technical stack: JavaScript runtime for strategies, Firebase backend, real-time market data ingestion. One piece I'm curious about: we built tt.ai() - strategies can call AI at runtime to make decisions (e.g., tt.ai("Dip or falling knife?", data)). Curious what the HN community thinks about AI making real-time trading decisions.

Happy to answer questions about the architecture, regulatory stuff, or how we're thinking about AI making trading decisions.