We've re-written the backend from scratch to allow each company to build their own data model. This lays robust foundations for one of our mid-term goals which is to establish Twenty as the most flexible CRM on the market. Thanks to the 33 contributors of this release 🙌
We released v0.20 with 121 improvements.
Here are some of them:
- audit logs for every object
- a new timeline design for every object that features audit logs, emails, calendar events...
- ability to create objects directly from an object page in a side panel (e.g. create a
Major perf improvements from the latest release!
No tool has helped us improve our product quality as much as
@getsentry
. Beyond bug tracking, it's also deep insights with anonymized session replays and performance profiling.
Thanks for supporting open source 🙌
Twenty v0.1.4 is out 🚀 We are happy to release two new features: Views and Board stages edition.
A huge thanks to all the community contributors (cc
@adityapimpalkar
and the 23 others!). Feel free to join us on discord:
We just released a few improvements:
- 5x frontend performance increase on list views - it should feel much faster when you have lots of records
- new custom field types: Multi-select, JSON, Date + Date & Time, Currency
Thanks to the 30 contributors who worked on v0.10 🙌
This is the kind of issue that would rot in a PM's backlog at most large companies because nobody gets promoted for doing small fixes.
Open-source wins because it enables swift iteration — from user request to feature built by a first-time contributor in less than 24 hours.
Really excited by
@thomasdesfrancs
's design concepts. Some amazing features coming to
@twentycrm
in H1. This will be 100% Open Open Source, free to self-host and available on the cloud hosted version at a much lower price than the market.
@flxmgdnz
Starting to work on it on Monday and it should take a month to ship it, we'll have it in prod by end of November for sure. Also working on custom objects (Oct 30) and email integration next! Hopefully we'll have passkeys one day too 👀
We were busy shipping and forgot to post about last week's release.
There were 169 PRs from 35 contributors. As always, most are small quality of life improvements. Here are a few you might notice:
@peer_rich
Amassing an open-source army of 200 here — looking at you,
@supabase
. Sorry
@kiwicopple
, pg_graphql's great and Twenty wouldn't work without it, but Peer was first check 🤷♂️
A new custom field type to add multiple links.
We haven’t migrated the company’s domain name to that field yet but that should come.
Soon, we also plan to allow multiple phone numbers and emails in the same way.
@BaptisteBCA
Salut Baptiste, merci beaucoup pour la vidéo ! On se regardera le replay en entier demain pour prendre tous tes retours - j'ai noté plusieurs point intéressants déjà. On a parcouru du chemin en un an mais il reste encore beaucoup à faire. Tu verras qu'on avance vite :)
@charlestehio
@romaindewolff
Thanks Charles! FYI we'll start working on Calendar integration (+ probably a meeting recording integration with soon after). Shipping beginning of April
Also from last week’s release:
- A new user-guide
- The ability to “click to copy” on fields
- Batch endpoints and improved error messages on the REST API
@JayGemini4
Yes it is! The trickiest part is the import, best is to import via the API as the CSV imported is limited to 2000 records per import for now (will change)
@ianuragbanerjee
@calcom
What we do on the backend is quite complex and not standard because we need to accommodate for custom objects for each workspace and have 1 schema for each. A lot to learn in our code base but we wouldn't recommend adopting the same pattern for a side project 😁
@getsentry
Some requests had become extremely slow. This was due to the way we compute the unique metadata schema for each workspace, which is now cached.