In developer tooling, if a "rough edge" makes a tool hard to use, but a "sharp edge" makes a tool easy to accidentally misuse, then what kind of edge should a tool have?
In developer tooling, if a "rough edge" makes a tool hard to use, but a "sharp edge" makes a tool easy to accidentally misuse, then what kind of edge should a tool have?
It was super fun chatting with @kball.llc about spec-driven development and the future of coding. Since recording, Kiro announced new types of spec flows. Last week I used Kiro's new Bugfix Spec flow to one-shot a bugfix to production! Very thorough tests, clear implementation.
AWS DevOps Agent is out! Try it out or learn about how it speeds up your operational investigation and improves your operational posture in the background: www.youtube.com/watch?v=JajB... aws.amazon.com/devops-agent/
#AWS #AWSreInvent #DevOps. Works with whatever #observability tools you use!
@kiro.dev
just went GA, and with that, new features! My favorite is Property-based testing. PBT takes your Kiro spec to figure out behaviors want the system to have, and uses those to generate thorough tests.
@devops.com.web.brid.gy devops.com/aws-extends-...
Spec-driven development with an #Agentic IDE like @kiro.dev can help keep it on task and remember key implementation goals like security features, while keeping the flow you get with vibecoding. Watch the chat with @ashimmy.bsky.social at @techstronggroup.bsky.social: techstrong.tv/videos/inter...
kiro.dev twitch stream at twitch.tv/kirodotdev
Come learn about @kiro.dev, a brand-new AI IDE that brings the fun of vibecoding into the scale of real-world production-scale projects. Recorded livestream here: www.youtube.com/watch?v=sXbI...
And of course check out the blog: kiro.dev/blog/introdu...
A new way to vibecode from idea, to spec, code, docs, tests, and everything else to deploy to production.
I've been using it for the last couple months and it has made it so that I can check things off the parts of my backlog that I never would have been able to get to!
kiro.dev/blog/introdu...
Great article by Greg Nudelman on Agentic UX: www.uxforai.com/p/secrets-of... . It covers Amazon Q Developer's new operational investigations in Amazon CloudWatch, and talks about how humans and agents collaborate toward ruling in and out possible causes. #AI #Observability #Operations #UX
New Amazon Builders' Library article: Resilience lessons from the lunch rush. Fantastic overview of techniques to prevent overload in distributed systems, and highly entertaining real-world stories that draw parallels. aws.amazon.com/builders-lib...
Recording available: And itβs live: youtu.be/ad42UTjP7ds
Apparently this year I started a tradition of wearing my slide art printed on my t-shirt.
COP322 βDonβt get stuck: How connected telemetry keeps you moving forwardβ. The recording is now available: youtu.be/ad42UTjP7ds #AWSreInvent #Observability
Come learn cognitive techniques and tools for reducing time to resolution. We'll talk instrumentation, logs, telemetry correlation, and I'll show how Amazon Q Developer for troubleshooting operational incidents helps you move toward resolution even faster. Thursday 2pm! #AWSReInvent #observability
When I built demos for my #awsreinvent talk about troubleshooting, I was blown away at how quickly Amazon Q Developer was able to root cause the failures I was injecting. Come learn more for COP322 on Thursday 3pm. Try the new troubleshooting capability in CloudWatch: aws.amazon.com/blogs/aws/in...
But to also know that it'll be recorded along with all my other attributes that I wrote down in the span so that I can slice and dice things in any way I want later on. So in short - yeah agreed about the wide event idea! And I do appreciate finding names to call these practices, so thanks for that!
I've always found it handy to be able to add some instrumentation in my code, and know that it'll show up as a dashboardable/alarmable metric with a predictable dimensionality (to balance cost with performance).
New blog post, with my personal take on our launch of Amazon Aurora DSQL: brooker.co.za/blog/2024/12...
Come nerd out with me about #Observability at #aws #reInvent! I'll demo lots of new features, and will also share all my secrets about how to troubleshoot and fix issues as quickly as possible.
Still seats left for the last showing of COP343 - Building #observability for improved #resiliency. 5pm in the Venetian. Another session repeat was added due to demand! #AWSreInvent
Due to popular demand, new session repeat just opened up! COP343 - Building observability to increase resiliency. Tuesday Nov 28 5pm. Should be lots of fun!
πππ New CloudWatch Logs "pattern" feature!! aws.amazon.com/about-aws/whatβ¦
Uses AWS ML to automatically recognize patterns in log data, aggregate related logs and summarize thousands of log lines into a few easy to visualize groupings. Find the needle in the haystack of log lines!