Quant On A Budget

How This Came Together

I didn’t set out to build a trading platform for the masses.
I just needed something that would reliably do a few specific jobs for me.

• I needed a scanner that could surface candidates without constant babysitting.
• I needed a watchlist that stayed current.
• I needed a clean way to turn setups into playbooks so I wasn’t reinventing the plan every time a name lit up.
• I needed alerts — on breakouts and on the exact conditions in those playbooks — delivered where I actually look (Telegram).
• And I needed to be able to see the whole picture from a browser when I wanted context.

So I started putting the pieces on a DigitalOcean droplet running the most stable Python version I could lock down. Interactive Brokers Gateway handled the market data and eventual execution side. A lightweight web server gave me a view into the system without having to SSH in every time. Telegram became the notification layer.

None of it arrived fully formed.

The real advantage was being able to build one small, working piece at a time — scanner first, then watchlist logic, then playbook structure, then the alerting rules, then the web view — and wire them together as they proved themselves.

Grok made that incremental process fast enough to stay in flow instead of getting lost in architecture diagrams.

Does an Off-the-Shelf System Exist?

Commercial tools exist for individual pieces—like Trade Ideas or TrendSpider for scanning, or TradersPost for webhook execution. But no off-the-shelf commercial product tightly connects this specific end-to-end workflow without forcing you into a rigid walled garden.

To cobble together commercial SaaS software to handle custom scanning, real-time data feeds, webhook bridges, and cloud hosting, you are easily looking at $150 to $300+ per month in recurring software subscriptions—and you still don't truly own your underlying infrastructure or stateful playbook logic.

By hosting a lean, custom Python stack on a dedicated droplet, operational overhead drops to around $10 to $20 per month (primarily just server hosting and raw market data costs). More importantly, you retain 100% control over every signal rule, database entry, and notification payload.

What exists now is not a fully polished product. It’s a working system that does exactly what I needed, stays under my control, and can still grow one focused piece at a time. It is the foundation for someone else to build their own.

---

I’m sharing the approach for the few who want to learn how to build something like this themselves.