How CRISPY is Struct?

In this post, I'll assess Struct from the lens of CRISPY framework that I had introduced in my June blog post.

Contextual Clarity: 1/1

We use AI for generating title, summary for every thread. Moreover, you can assign multiple tags to attach additional context. You can even convert Struct into a task management system, like we've done on our team.

Real-Time: 1/1

Struct's feed is IMO the most real-time feed there is, adding, updating and removing threads live. Struct uses Websockets heavily to bring anything happening anywhere, real-time to the users.

Isolation: 1/1

Struct's Thread based design allows conversations to evolve without interleaving. Moreover, if a conversation in a thread goes out-of-scope, you can "fork" those chat messages into a new thread. Or, merge two very similar threads together (coming soon).

Spam Control: 0/1

Coming soon. We intend to build multiple systems to deal with spam, including trust-levels and AI-powered moderation.

Progressive Learning: 0/1

Struct Bot can act like a GPT over your past accessible conversations, and respond directly to your queries.

Yieldful Search: 1/1

We use hybrid search, a mix of keyword and semantic search, to search the most relevant threads for your search query. Moreover, when you're creating new threads, we propose to you similar threads that might already have what you're looking for. We also show similar threads for every thread that you're looking at. 1/1

As of Feb 22, 2024, Struct scores a 4/6 in the CRISPY framework.