TIL web neural network API, an emerging web standard for building in-browser AI apps that lets you use those new NPUs in phones and laptops. w3.org/TR/webnn

I have not paid a ton of attention to the uproar over RTO policies, bc we are all in on distributed teams and not going back. My impression (via social media) has been that these were shadow layoffs. Last month I asked an investor why they are doing RTO. He said: "Retention, mostly. And morale."

I've been diving into Neon and I like what I see! Here's how you can use Neon with ephemeral environments to simplify serverless development. theburningmonk.com/2025/04/how-...

Saying "coding is dead" is missing the point. Coding isn’t just about producing code, it’s also a way of thinking. It teaches us structure, systems thinking, abstractions and problem decomposition. If you don’t learn those skills, AI won’t make you productive, it’ll make you dependent.

I finally binged The Orville, it was time well spent, especially seasons 2 and 3! So high on Star Trek nostalgia right now 😅

I recently spoke with Waldemar Hummer to find out what's new in LocalStack v4 - e.g. visualize and trace event-driven arch end-to-end, or create snapshots of your system. It's crazy how fast they are innovating! youtu.be/vVUBnXD6eto

:-P

Through chaos experiments, the Lambda team reduced response time to an AZ outage by 90%! 🤯

"Lambdaliths have slower cold starts" "Lambdaliths have fewer cold starts" Both can be true, but neither is entirely correct. It depends on several factors. In my new post, I explain these nuances so you can make a better choice. theburningmonk.com/2025/03/the-...

I spoke with Khawajas Shams about building large-scale (think millions of TPS) multi-tenant systems. We talked about pros & cons of cell-based architectures, how it reduces the blast radius, and techniques for building high-throughput systems e.g. core pinning www.youtube.com/watch?v=8jte...

Nice to see a full house at the serverless days Manchester 💪

Thank you to everyone who attended my talk at the Serverless London meetup last night 🙏 Here are my slides for those who couldn't make it. You can try Evolua for free at evolua.io. We'd appreciate your feedback! speakerdeck.com/theburningmo...

New post: How we built an AI code reviewer with serverless and Bedrock My biggest lesson learnt is that, at least for our use case, LLM is the easy part and there's still a lot of "app" in an "AI-powered app"!

Wiz found a publicly accessible ClickHouse database linked to DeepSeek, completely open and unauthenticated, exposing sensitive data like chat history 🤦‍♂️ www.wiz.io/blog/wiz-res...

I spoke with the co-founder of @hookdeck.com to find out what it can do, and it's impressive! There is a lot of focus on DX for building Event-Driven Architecture, and EventBridge can learn a lot from them. Have a look for yourself. www.youtube.com/watch?v=-jmM...

AWS Hero status renewed for another two years 😇 aws.amazon.com/developer/co...

This is AWESOME!! EventBridge now delivers events to cross-account targets directly without having to send them to the default bus in the target account first. More details here: aws.amazon.com/about-aws/wh...

Here's a mindmap of everything we'll cover in my 4-week boot camp starting on Monday. Are we missing anything that one should know for running serverless workloads in production? Detailed curriculum here: productionreadyserverless.com#curriculum

We're back with a new episode of real-world serverless. I spoke with @boyney123.bsky.social about the hard parts of building event-driven architectures and how EventCatalog can help with event documentation, discovery, versioning, visualization and testing. www.youtube.com/watch?v=ZVl7...

oh, this is a toughie...

Stop wasting money on CORS preflight requests. Here's a quick tip for reducing API Gateway costs. 👇

Confessions of a Big Tech and startups recruiter: with Blake Stockman. Blake has been a tech recruiter for 12 years, working at Google, Meta, Uber, Flexport, Y Combinator and startups. Full episode: • Spotify: open.spotify.com/episode/6sA6... • YouTube: youtu.be/d7GsnaQYZ68 (cont'd)

This is still my favourite answer to "What does a software architect do?" Source: youtube.com/watch?v=FGKG...

Current status: working on Terraform track for productionreadyserverless.com ✅ Set up TF project & serverless.tf ✅ Support multi-account ✅ REST APIs with API Gateway and Lambda ✅ Testing with ephemeral environments ⃣ Event-Driven Architectures ⃣ Observability & performance

I'm speaking at AWS Community Day Romania this April, in the beautiful city of Timisoara. Hope to see you there 👋 Their CFP is open until February, so there's still time to submit your talk. More details here: aws-community.ro

You might be paying for unused WAF ACLs that you can't see. Here's what I found in my latest bill. 👇

2024 was the year I got healthy 💪 and got back amongst the community. And it felt great! Here's what I got up to in 2024 and my most popular content: theburningmonk.com/2024/12/year... Wish you all a happy new year and hope to catch you in 2025! Ciao for now 👋

I just burned $10,000 on marketing, and it taught me the hardest lesson of my career. I spent $10,000 on an ad agency and only got 3 sign-ups for my workshop, despite months of ad campaigns and plenty of clicks. Yet, 30% of students joined through word of mouth, another 30% through social media.

Do you build mulit-region applications on AWS? Then bookmark this page of AWS services and their multi-region capabilities. repost.aws/articles/AR0...

An eng manager at Netflix once told me that "It'd cost 7x to run Netflix if Netflix ran on Lambda". But what he said next was even more insightful. "But, not everything at Netflix runs at Netflix scale." And that's true for most enterprises!

When you use fetch in Node 20 or later, the X-Ray SDK doesn't propagate trace headers downstream. If you have a Lambda -> API Gateway -> Lambda, the two invocations will not appear in the same trace. This is the two-line fix you need. (you need to install the aws-xray-sdk-fetch module first)

Hey look at that... one of @theburningmonk.com's top podcast episodes of 2024 was about LocalStack. Maybe we should take a look... (it's at v4 these days). www.youtube.com/watch?v=BP2U...

"I just got disowned 'cause you made him believe I'm a Java dev!" 😂 😂 😂 www.youtube.com/shorts/TwXu4...

Current status: refreshing content for my Jan 2025 workshop ✅ Update Serverless Framework track to v4 ✅ Update to ESM modules ✅ Update SSM vs. Secrets Manager comparison ✅ Jest -> Vitest ✅ Update SNS vs. SQS vs. EventBridge vs. Kinesis comparison ✅ More details on remocal testing

[New Post] How to end-to-end test microservices across bounded contexts? Here, I discussed the pros and cons of several approaches and explained why I prefer using mock APIs instead of hosting other teams' microservices. Enjoy! theburningmonk.com/2024/12/how-...

Skype is pushing everyone to a subscription model for voip phone calls. Are there any alternatives that's still pay-as-you-go? I looked at a few, like Viber, but they are all subscriptions-based as well.

There is still coupling in Event-Driven Architectures (EDAs)! Coupling means changes on one side have to propagate to the other side and it can be measured in multiple dimensions. Even in EDAs, there are multiple forms of coupling:

Serverless Picks of the Week #142! Our serverless🦸‍♂️is Philipp Krenn, DA and community lead at Elastic! Philipp really understands the hard part of serverless data, and does a🔥job explaining it to others. Thank you for everything you do, Philipp 💙! www.readysetcloud.io/newsletter/142

This is easily one of the most common questions I get about Cognito. Here's how to invalidate Cognito-issued JWT tokens (including a working demo)

I've uploaded a new video on CQRS (Command Query Responsibility Segregation), which explains what it is, how to implement the pattern, when to use it, and its pros & cons. Watch it here: https://buff.ly/4giqGqH I'd love your feedback in the comments!

Hackers stole over two terabytes of sensitive data, but they put the data in an open S3 bucket and that's how they got caught. Oh, the irony... hackread.com/shinyhunters...

Excellent guide on writing useful error messages: * Use the correct status codes * Be descriptive * Be helpful * Provide elaborative information * Be empathetic Read the full article for some examples:

ICYMI, my next serverless workshop is going to include a new Terraform track. That's 3 courses for the price of 1! 👉 https://buff.ly/2JVuG1o ps. early bird (30% off) tickets are available until Dec 16th

Finally home from re:Invent! Long journey, but me, Sam and Tim all made it back in one piece.

In this video, I explain the what, why and when of Event Sourcing and the problems you might encounter. Hope you like it!

Here are the biggest re:Invent serverless announcements theburningmonk.com/2024/12/bigg...

re:Invent 2024 has been amazing, but... www.youtube.com/watch?v=XLmU...

@nickt.com.au you didn't tell me you were on national TV! found this while looking for a video of your lights www.youtube.com/watch?v=fsL2...

Pre-keynote Hero breakfast at the HashiCorp lounge! #reinvent @theburningmonk.com @richardfan.xyz @linda-mhmd.bsky.social @anahit.bsky.social 🥰🥰🥰🥰

Thank you @antstack.com for the thoughtful gift! It's just what I needed 😀