You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been using CALM for my own architectures at home as I build out my network.
One of the things I have been doing to significantly speedup creating my specifications is to use ChatGPT.
I have found it fantastically quick.
To achieve this I have passed in the Specifications to ChatGPT and then asked it to produce a model for me.
e.g. Can you create an architecture the connects a Grafana front end to a InfluxDB backend, Influx DB is pushed metrics from a BackendAPI
I've then asked it questions about performing threat modelling on the architecture, what are the security holes I have or where should I rate limit.
I have found it great to explore an architecture and generate them for me. Bonus is that the generated architectures are accurate as well.
I thought that maybe we should bring up this topic of using LLMs on architectures to generate and explore them.
Using ChatGPT with CALM
I've been using CALM for my own architectures at home as I build out my network.
One of the things I have been doing to significantly speedup creating my specifications is to use ChatGPT.
I have found it fantastically quick.
To achieve this I have passed in the Specifications to ChatGPT and then asked it to produce a model for me.
e.g. Can you create an architecture the connects a Grafana front end to a InfluxDB backend, Influx DB is pushed metrics from a BackendAPI
I've then asked it questions about performing threat modelling on the architecture, what are the security holes I have or where should I rate limit.
I have found it great to explore an architecture and generate them for me. Bonus is that the generated architectures are accurate as well.
I thought that maybe we should bring up this topic of using LLMs on architectures to generate and explore them.
TAG: @rocketstack-matt
The text was updated successfully, but these errors were encountered: