I built a local-to-cloud (AWS) bridge for Kite to solve the TradingView webhook issue. Looking for feedback on the architecture.

Hi everyone,

Like many here, I was incredibly frustrated with TradingView webhooks dropping, token expirations on Kite, and the risk of my local PC going to sleep while a trade was active.

Instead of relying on a fragile local Python listener, I ended up decoupling the whole system. I wanted to share the architecture I built and see how other algo traders here are handling this, or if you see any flaws in my logic:

1. The Local Desktop UI: I built a local interface where I log in and set my strike parameters, risk thresholds, and triggers.

2. The AWS “Brain”: When I click “Arm,” the UI pushes those parameters to a dedicated AWS server.

3. Cloud Execution: The AWS server handles the 24/7 Kite Connect streaming and calculations. If a condition is met, the cloud executes the trade directly via the API. Even if my laptop dies or loses Wi-Fi, the cloud brain manages the exit.

4. Live Sync: The execution status pushes back to my local UI so I can monitor it safely.

Has anyone else moved their Kite execution engine entirely to the cloud while keeping the control panel local? Would love to hear what stacks you are using (I’m using Python/FastAPI on the backend) and how you handle latency between the cloud and Kite’s servers.

This is easily mitigated by setting up power management rules. Set it to not go to sleep ever.

Never had this issue. Once a day is what is needed


I gave up on AWS after a scare recently where a bug on their side showed gigantic $ usage. Something like this, there are lots of such threads. Was pretty scare initially.

That was a bug, but AWS has this risk of unlimited postpaid bill that we cannot control. Risk chance may be low with good security like yubikey, but if by chance your pc/account gets hacked damage can be severe.

And AWS support for retail seems to suck, , no phone number. Fill a form. My experience made it clear that it wasnt worth it for me.