Ryan Delaney

1.3K posts

Ryan Delaney banner
Ryan Delaney

Ryan Delaney

@_rrdelaney

Coding agents @Linear Prev: Co-founder @RevereCRE, Closure Compiler @Google, DevX @Meta

Sunnyvale, CA Beigetreten Ocak 2017
520 Folgt771 Follower
Ryan Delaney retweetet
Linear
Linear@linear·
Following the US government's export control directive and Anthropic’s recent announcement, we are suspending access to Claude Fable 5 in Linear Agent’s coding sessions. To ensure minimal disruption, all workspaces who have enabled Claude Fable 5 are being automatically failed over to Claude Opus 4.8. We will provide updates as they become available.
Anthropic@AnthropicAI

The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Claude models is not affected. We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible. Read our full statement: anthropic.com/news/fable-myt…

English
2
7
222
30.8K
Ryan Delaney
Ryan Delaney@_rrdelaney·
@mingjie Merging my pending Fable-authored PRs is going to feel like living in the future
English
1
0
5
246
Mingjie Jiang
Mingjie Jiang@mingjie·
1) what
ClaudeDevs@ClaudeDevs

As a result of a US government directive, we are suspending access to Claude Fable 5 for all users. You can continue to use all other Claude models. Here’s what this means for you: Across Claude products, new sessions will run on your selected default model or Opus 4.8, and existing Fable 5 sessions will end with an error. On the Claude Platform, requests to Fable 5 will also return an error. Please update your integrations to other Claude models. We know this is a disruption to your workflows; we appreciate your patience and support.

English
4
1
44
7.7K
joe feeney
joe feeney@iamsomewalrus·
@karrisaarinen @linear Do you have documentation the setup eg using dev containers to specify environment, etc? Can the agent use chrome or something similar to QA / test
English
1
0
0
269
Ryan Delaney
Ryan Delaney@_rrdelaney·
@jhasanofficial @karrisaarinen @linear Coding sessions run in a secure sandbox with no access to secrets or credentials, and limited GitHub access. Additionally, for externally created issues we lock down the sandbox's network access.
English
1
0
1
59
Jahid
Jahid@jhasanofficial·
@karrisaarinen @linear That's a lot. Were there any complex bugs it was able to fix, or were they just small PRs? Also, how do you handle a linear agent not adding malicious packages when it suggests code or pkgs to use? Any kind of proactive measurement?
English
1
0
0
134
Michelle Mayes
Michelle Mayes@michellejmayes·
@linear It auto starts sessions, doesn’t give me a chance to reply if it auto picks the wrong one. Generally commenting the repo before assigning to Linear prevents this though.
English
1
0
0
34
Michelle Mayes
Michelle Mayes@michellejmayes·
@linear I like coding agent sessions but sometimes it randomly picks the wrong repo and there's nothing I can do to fix it. The session and credits are gone. How can I deterministically make it pick the right repo without adding a comment "repo: X" ?
English
1
0
1
48
Ryan Delaney retweetet
Kenneth Skovhus
Kenneth Skovhus@kenneth_skovhus·
Imagine having all your company’s context in one system. Then teaching that system how to code. 💥 Issue → Code → Review → Land All from the same interface. Already, 30% of our triage issues are automatically fixed by the agent. And we are just getting started.
Linear@linear

Introducing coding sessions. Linear Agent can now triage issues, investigate the cause, write the fix, open a PR, and bring the code back for review. All shared with your team in Linear.

English
0
1
13
1.3K
Ryan Delaney
Ryan Delaney@_rrdelaney·
@oliverbeavers Woah that’s super cool! Yea we used Linear at Revere and I liked it so much I figured I should help build it
English
1
0
1
63
Oliver Beavers
Oliver Beavers@oliverbeavers·
@_rrdelaney Interesting to see you on this from Revere! We shared an investor.
English
1
0
0
68
Ryan Delaney
Ryan Delaney@_rrdelaney·
I’ve been working on this at Linear for the past few months. You can now tell the Linear agent to create and work on an issue, or manually delegate to it yourself. Coding is done within a secure sandbox, which spins up fast. Ask for follow-ups inside of diffs, all without leaving Linear. Lots of improvements to come, let me know what you want to see :)
Linear@linear

Introducing coding sessions. Linear Agent can now triage issues, investigate the cause, write the fix, open a PR, and bring the code back for review. All shared with your team in Linear.

English
4
1
33
6.4K
Thakur Kharel
Thakur Kharel@takurooman·
@linear and @karrisaarinen how do I opt out of coding ? I get hit with this every time i tag the bot in slack. We are not ready yet to adopt this :).
Thakur Kharel tweet media
English
1
0
0
51
Ryan Delaney
Ryan Delaney@_rrdelaney·
Sandboxes aren't VMs, but you still probably want an init system and process monitor for the services you're running inside of them. Recommend checking out dinit!
English
0
0
4
96
Ryan Delaney
Ryan Delaney@_rrdelaney·
@kentcdodds Why is this an issue? Multiple lazy loaded queries can fire off after an initial page load, each one not over-fetching. Or like, just pagination looks like this.
English
0
0
4
546
Kent C. Dodds 🏹
Kent C. Dodds 🏹@kentcdodds·
When you see this in your network tab on page load it makes you wonder why they chose graphql...
Kent C. Dodds 🏹 tweet media
English
59
15
844
120.8K
Ryan Delaney
Ryan Delaney@_rrdelaney·
With all this talk of sync engines coming back up, quick reminder that the best open-source sync engine is still Relay, and the team just shipped a major new version! And small plug, but the best way to build Relay apps is still Pastoria, which I'll be updating this weekend
English
1
0
1
176
Ryan Delaney
Ryan Delaney@_rrdelaney·
@ToastingTweets We’re trying out a couple of models and harnesses to find the best fit, and I expect it will change post launch, especially as we look into more performance improvements
English
1
0
0
46
Toast
Toast@ToastingTweets·
@_rrdelaney What model and thinking level does Linear Agent use?
English
2
0
1
56
Ryan Delaney
Ryan Delaney@_rrdelaney·
@olvrgln @linear Appreciate the feedback I think you'll get early access to our in-house coding agent soon which should address some of these problems. We're not at full feature parity with the other cloud coding agents yet, but the simple delegation workflow should be super smooth
English
1
0
3
52
Oliver
Oliver@olvrgln·
Ultimately there's too much config that lives in the external agent. How can I choose a model or thinking level for this specific ticket, how can i adjust the behavior of how the agent creates a pr vs doesn't, etc. Strange too that the PRs created by Cursor for example from a linear assignment doesn't automatically link the PR to the ticket! Ultimately I think what I want/need is twofold: - one place to manage, triage, and delegate the work to be done and how it should be done - an agent that has context to my linear and the rest of the systems. These agents are repeatedly putting up PRs that don't pass CI. They should basically just keep working until CI definitely passes without me telling them
English
1
0
1
230
Oliver
Oliver@olvrgln·
trying a bunch of cloud agents with @linear and so many papercuts. Still very early to create a good experience here. Also finding that linear isn't really as good a command center to delegate to these agents as I want. May writeup my thoughts here
English
7
1
31
7.7K
Ryan Delaney
Ryan Delaney@_rrdelaney·
I’ve written more C at Linear than all my previous jobs combined
English
1
0
1
170
Ryan Delaney
Ryan Delaney@_rrdelaney·
I really wish the tsgo LSP worked better in Zed, but it's either constantly crashing or not even type-checking the code I write. Back to VS Code 🥲
English
1
0
1
721
Ryan Delaney
Ryan Delaney@_rrdelaney·
As someone recently impacted by this at Linear, this was tough to read 😔
Tuomas Artman@artman

Today is a hard day. I shared this note with the @linear team today: We’ve made the difficult decision to increase our workforce. This is not a cost-cutting exercise or a reflection of anyone’s performance. We’re simply reimagining every role for the agentic AI era. We’re hiring. We’re sorry about that.

English
1
0
45
12.7K

Entdecken

@mingjie @iamsomewalrus @karrisaarinen @linear @jhasanofficial @michellejmayes @oliverbeavers @takurooman