aleeve.dev
Scalable and distributed custom feed generator, ott - on that topic
Fix really bad spelling and wording..
+4
-5
README.md
+4
-5
README.md
···
5
5
The flow is as follows:
6
6
7
7
1. Two fluvio connector services consume wss streams, one for raw-posts and one for raw-likes
8
-
2. ott-filter consumes the keyed posts, likes stream and keeps count on likes and other filters,
9
-
upon pass it sends to passing posts to the fluvio topic posts.
10
-
3. ott-embed consumes the posts topic, embeds them with tei running on hosts and stores the vectors in a pg cluster
11
-
4. ott-xrpc listens to getFeedSkeleton requests, gets the users last like and does a vector gets similar posts from the pg db.
8
+
2. ott-filter consumes the keyed posts, and likes streams and keeps count on likes and other filters.
9
+
It sends the passing posts to the fluvio topic posts.
10
+
3. ott-embed consumes the posts topic, embeds them with tei running on host and stores the vectors in a pg cluster
11
+
4. ott-xrpc listens to getFeedSkeleton requests, gets the users last likeed post and gets similar posts from the pg db.
12
12
13
13
Still work in progress, especially the ott-xrpc service isn't fleshed out yet. Also I intend to add a VIP stream so that all posts
14
14
liked by a feed user are guaranteed to pass the filter.
15
-
16
15
17
16
Then... The original intention was to use FASTopic to get topic vectors for every day in the semantic space and sample within the most relevant topics.
18
17
This is still the intention.