Keith Williams

Just want to say to all the engineers who are more “heads down” and focused on delivering high quality software instead of trying to build whatever will get you attention, keep doing your thing. Our industry needs more engineers like that. 🙏🏼

AI being able to regurgitate code that has been written similarly hundreds or thousands of times means you can use your newly found time to be creative and to truly innovate. That should be exciting, not scary.

@shashanktelkha1 Awesome. Please apply using the link I posted

Hi Keith — I work as a Software Engineer at Capital One, where I focused on scaling distributed systems handling millions of records and transactions daily. I’ve done deep performance investigations, caching strategies, and reliability improvements in production-scale systems. This IC4 Scale role sounds right up my alley — would love to apply.

I'm hiring an IC4 Scale Engineer. If you are passionate about scaling DBs with millions of records, global deployments, sophisticated caching techniques and tons of perf investigations, would love to see your application. cal-com.breezy.hr/p/52678610e5f1…

@dhaiiryashiil @calcom @peer_rich Damn I get a VIP badge. Nice 😀

Happy to announce that I will be joining @calcom as a full time engineer Pumped up to give everyone the best experience to schedule your meetings! Thanks to everyone involved, LFG 🚀

@teodos @calcom Hey @teodos want to DM me your username and I’ll foreword to our support team

@keithwillcode cc @calcom ^

On Tuesday, we regrettably had degraded service on Cal.com with very long load times and even downtime for some users. Here's a write-up of what happened, how we fixed it and what we did to improve. PRs included. cal.com/blog/incident-…

Getting the green line as close to the purple line as we can. Huge influx of community PRs over the last 9 months that are we so grateful for and we want to keep up. Changes coming...

@hbjORbj @Lovable @GoDaddy @vercel Awesome 🙏🏼

I gift people around me with custom websites all the time. ex) my boxing coach, local gym, dad, friends, and more The hour I spend usually goes a long way for them. I recommend you do the same if it's easy for you but not as easy for them I use @Lovable , @GoDaddy , and @vercel

@hugodemenez @peer_rich Thanks for reporting. We'll get it fixed cc @dhaiiryashiil

@keithwillcode @peer_rich I reported this bug in TestFlight (state doesn’t update on dates overrides).

Built an automation where adding labels to PRs has @DevinAI resolve conflicts and even push stale PRs across the finish line for you. Very curious how this accelerates our core team and especially the community. Let me know of any feedback you have. 🙏🏼

Technology that results in people critically thinking less is not ideal, so we need to be careful with AI. To combat this, I’ve decided to leverage AI to reduce the wasted time on syntax, quirks, etc. of coding and instead unleash a flood of ideas.

Had fun with this PR just now. I forgot that GitHub doesn't wait for pull_request_review to be in main and I saw changes coming in from "nowhere". Then realized the AI inception had taken place and Devin was already fixing Cubic's comments. github.com/calcom/cal.com…

@JayChopra_ @lottsnomad 🙏🏼

@lottsnomad Trust me on this cal is sm better, also shoutout @keithwillcode haha

is calendly the best ? what else would you suggest? im doing a lot of customer calls for context would also embed it into our site too

@MEbandhan @peer_rich Partially. But yes, @iamvolnei has been doing a lot of clean up that would allow us to swap if desired. We just chatted about it yesterday because we know the speed benefits there for dependencies.

@keithwillcode unrelated but I have a qstn. Have you thought of moving to bun in future? I remember once @peer_rich mentioned that yarn is good for workspaces. Is that the reason that Cal is still using yarn?

Starting 2026 with an 83% reduction in CI wall times. Read the blog here -> cal.com/blog/how-we-cu…

Changed our GitHub actions to @useblacksmith last night. - 2x faster cache download speeds - unlimited vCPU concurrency - 33% faster PR workflows Impressive work by their team