Brian Scanlan
5.3K posts
Brian Scanlan
@brian_scanlan
I work at Intercom.
Swords, Dublin, Ireland เข้าร่วม Mayıs 2009
2K กำลังติดตาม4.9K ผู้ติดตาม
If you are interested in learning how to teach an LLM to review code like a senior engineer, I strongly recommend watching this video from my colleague @gregolsent (link in thread)
English
Brian Scanlan รีทวีตแล้ว
There is, of course, one AI Agent for Customer Support with public docs, self serve sign-up, public pricing, a CLI, API + more.
It's the highest performing one, and we share all our ideas+research too.
Product → fin .ai
Research → fin .ai/research
Ideas → ideas.fin .ai
Brendan Falk@BrendanFalk
I love how all the major AI customer support platforms are still so secretive about their API docs. It's May 2026 guys. Your agents use tools, skills, and system prompts like everyone else...
English
@mitsuhiko DNSSEC: Pointlessly replacing a minor security concern with a massive availability risk since 2004.
English
Now that we know it's terrible, can we finally kill DNSSEC?
English
Brian Scanlan รีทวีตแล้ว
This is a *must watch*: it was a pretty killer meetup (just check the open mic sessions!) and Intercom’s Claude Code Plugin system was the center of it. Every team building with Rails and agents needs to watch.
Evil Martians@evilmartians
April's SF Ruby Meetup is now live on YouTube! This time, we met at @intercom and listened to: - Brian Scanlan (@brian_scanlan) on turning Claude into a full-stack engineering platform - Konstantin Gredeskoul (@kig) on adaptive user flows with Ruby DSLs and LLMs - Chris Davis (@tweetellington) on semi-autonomous software: putting the human in the loop with Rubot Hosted by @inazarova, we meet every month to learn from each other and listen to presentations about all things Ruby (e.g., libraries, frameworks, open source tools, new startups). Watch the recording: youtu.be/xep8UoK5cyA?si…
English
Brian Scanlan รีทวีตแล้ว
April's SF Ruby Meetup is now live on YouTube! This time, we met at @intercom and listened to:
- Brian Scanlan (@brian_scanlan) on turning Claude into a full-stack engineering platform
- Konstantin Gredeskoul (@kig) on adaptive user flows with Ruby DSLs and LLMs
- Chris Davis (@tweetellington) on semi-autonomous software: putting the human in the loop with Rubot
Hosted by @inazarova, we meet every month to learn from each other and listen to presentations about all things Ruby (e.g., libraries, frameworks, open source tools, new startups).
Watch the recording: youtu.be/xep8UoK5cyA?si…
YouTube
English
@annimaniac Excellent framing. The maturity ladder is very useful.
English
If anyone is creating a company that helps companies that pulls companies up these levels, I'd love to meet. Particularly around legibility and/or non-engineer usage of AI
Ann Miura-Ko 🦖@annimaniac
English
English
intercom-client has also been popped by TeamPCP:
npmjs.com/package/interc…
@intercom
English
If the community wants to save contemporary GitHub we should rally around @kdaigle becoming independent CEO. He has been there since the old days. Knows the soul of the product and knows how to do it.
English
I found my lost coffee mug thanks to Amazon Quick! It searched my calendar over the last month and put together a search plan for where to look! #AmazonQuick #QuickAF @awscloud
English
Within a few hours of work, and a week of running in "shadow" mode, we rolled out the new renderer and now we can tell you who's on-call up to 2,500x faster for complex schedules.
The full blog post can be found here: incident.io/blog/whos-on-c…
(7/7)
English
Brian Scanlan รีทวีตแล้ว
i finally wrote down everything that i learned about compensating sales on a product with usage-based pricing. it should be relevant for most AI companies. i made some mistakes so that you don't have to! meghangill.substack.com/p/the-sales-co…
English
I had a fun conversation with Ankit about how we’re getting incredible results from aggressively adopting AI in our R&D team. Check it out!
Ankit Jain@ankitxg
Fascinating conversation with @brian_scanlan about how @intercom really drives AI adoption at a 400-engineer scale. They set a goal. 2x throughput in one year. They put their highest-impact people on that. People who can get stuff done and make decisions. They updated job descriptions, anyone not contributing is no longer meeting expectations. They also did a lot of enablement, ran hackathons, and AI immersion days. Went all in on Claude Code and created 300 Skills. New episode🎙️link in reply
English
I wrote up some thoughts on why SaaS needs to become agent-first, and why we built a cli.
English
@artee_49 You can't delete an S3 bucket before emptying it. I think they are directionally correct.
English
Brian Scanlan รีทวีตแล้ว
We’ve seen workloads double in the past three months on @buildkite. Will be well on track to 10x at this pace on an already huge volume base. We run CI for the fastest moving companies like Shopify, Uber and most of the frontier AI labs with a similar workload profile in terms of stateful guarantees. The amount of code being pushed is truly unprecedented.
English
Wanted to provide more clarity about this.
Yesterday, we had a regression in merge queue behavior where, in some cases, squash or rebase commits were generated from the wrong base state, making earlier changes appear reverted in branch history. 2,804 pull requests out of over 4M merged on April 23 (roughly 0.07%) were affected. We fixed the issue, we've contacted every impacted customer, and we're expanding our automated test coverage for merge queue operations. The team will be updating the status page with RCA details as well.
Tom Elliott@theotherelliott
This GitHub incident is insane. Merge queue commits have been reverting previously merged commits at random. This not only breaks the mental contract teams have with Git in general, but is subtle enough to be really hard to unravel after the fact. githubstatus.com/incidents/zsg1…
English