#37 Model Driven Agents 🤖
This week, our Serverless expert is AWS Community Builder Matthew Wilson, our spotlight falls on AWS Lambda Powertools expert Andrea Amorosi, we look at the latest AWS service releases, news, & more!
Welcome
In last week’s issue, our serverless expert was AWS Community Builder Tycko Franklin, and our spotlight fell on AWS Community Builder Lee Priest!
This week, our serverless expert is AWS Community Builder Matthew Wilson, our spotlight falls on Senior Solutions Architect, Powertools for AWS, Andrea Amorosi, we look at the latest AWS service releases, blog posts, hints and tips, news and more!
This week’s newsletter is sponsored by Leighton.
📰 Articles that caught the eye
Here are some stand-out articles I read during the week in the World of Serverless!
⭐ My favourite this week was the article by Sebastian from epilot showing how flipping a complex issue on its head to a simple implementation is a win!
Ran Isenberg did a great article called “I Tried Running an MCP Server on AWS Lambda… Here's What Happened”.
Anton Martyniuk covers “Building Multimodel AI Chat Bot in .NET with ChatGPT and Database Branching in Neon Postgres“.
Sebastian has an interesting article on “Scaling Notification Systems: How a Single Timestamp Improved Our DynamoDB Performance“.
Vadym Kazulkin covers “Quarkus 3 application on AWS Lambda- Part 2 Reducing Lambda cold starts with Lambda SnapStart“.
Arpad Toth discusses “Triggering multiple workflows with DynamoDB Streams“.
Uriel Bitton covers “AWS Lambda Best Practices For Performant & Scalable Serverless Functions“.
🎓 Ask the Expert
Each week, I ask a different serverless expert the same three questions to get their personal insights - this week, we have AWS Community Builder Matthew Wilson:
Opinions are the author’s and do not express the views of their employer.
1. What is one common mistake you see teams making when implementing serverless solutions, and how can they avoid it?
I’ve been incredibly fortunate to work on a number of greenfield products over the past few years in which we adopted a serverless-first approach (simply using serverless technologies when it makes sense). From this context, where you are starting from an empty AWS account and GitHub repo, a common mistake I’ve seen is teams just jumping in to implementation without too much upfront education or planning.
With serverless it is incredibly simple to get something working by connecting a few services together but this can become very difficult to manage or scale as the project matures.
At an individual level my number one advice is to listen to the experts, this list below is not exhaustive but it is where I commonly go for inspiration:
https://blog.serverlessadvocate.com/ [Serverless Advocate Blog]
https://serverlessland.com/patterns [ServerlessLand Patterns]
https://theburningmonk.com/posts/ [Yan Cui Blogs]
https://www.instil.co/category/serverless/ [Instil Articles]
The goal here is not to necessarily memorise every bit of content but instead try to build up a “developers toolbox” of patterns, services and tools that can be used to solve a problem as well as any common pitfalls that other teams have fallen into. I appreciate that this can be incredibly hard to keep on top of as project demands can at times make personal development difficult but the best engineers that I’ve worked with are the ones that set aside time for themselves.
At a team level my best advice when starting from scratch is to work with your client / customer / business to determine what the ultimate end goal of the solution should be. The Value Flywheel Effect calls this “clarity of purpose.”
Typically at this point you will have a fair idea of what you need to achieve and probably a few solutions in your head on how you can solve the problem. The combination of clarity of purpose and a team of developers that have educated themselves on best practices will be a force to be reckoned with!
2. Which serverless tool or service are you most excited about right now, and why?
What I’m most excited about now is the intersection of AI and Serverless. We’re experiencing a huge demand from customers for how best to adopt Generative AI into their workloads and I believe that the engineers with strong serverless foundations are best placed to take advantage of the opportunity.
One great example is the concept of using AI agents for orchestration. Imagine the power of Step Function State Machines combined with the ability to adapt and work toward goals, rather than simply following well-defined states. I think this approach offers a number of advantages over what we get with Step Functions today, such as being able to choose the next best step in situations with ambiguity, or having greater resilience when a step fails. Step Functions are great for retrying, but adding a bit of intelligence to decide when to retry and when to fail would be amazing.
Something I’m particularly keen to explore is using Step Functions for well-defined workflows, exposing them as tools to an agent, and then letting the agent coordinate across the various State Machines it has access to.
3. What is your favourite trick or tip when working with serverless that the readers may find interesting?
When I first started developing serverless applications, the common guidance was to test everything in the cloud. “Bring the developer to the cloud, not the cloud to the developer” was a phrase I heard all the time. However, I’ve since learned that you really need a mix of both approaches.
Yan Cui spoke at the London Summit in 2024 about the idea of “remocal testing”. In my most recent project, we adopted this strategy by making it a part of our development lifecycle. We enable engineers to run entire services locally, while still connecting to remote resources like DynamoDB tables or S3 buckets in AWS.
Being able to debug code locally in your IDE, quickly make changes, and test them immediately is critical for maintaining good developer velocity and I’d highly encourage teams to think carefully before throwing that away for a “test in the cloud” approach and give remocal development a try.
✅ Bonus tip: join the hashtag#believeinsls discord! There is a community there to answer any questions you may have without getting overzealous on serverless or without judgment! Check it out!
🚀 New Releases
Here are the latest and most interesting releases this week in the AWS World:
⭐ The most interesting this week in my opinion is the release of AWS Strands!
Announcing migration assessment capabilities of AWS Transform.
Amazon CloudWatch RUM adds support for Interaction to Next Paint (INP) Web Vital.
Amazon SageMaker HyperPod now integrates with Amazon EventBridge to deliver status change events.
AWS CodePipeline now supports deploying to AWS Lambda with traffic shifting.
🔥 Tip: Check out https://aws-news.com/ for the very latest up-to-date serverless releases as they happen, created by the talented AWS Serverless Hero Luc van Donkersgoed.
👷🏻 Tools & Frameworks
Check out the latest open-source frameworks, news, and tool updates from the past week.
Log-Analyzer-with-MCP - A Model Context Protocol (MCP) server that provides AI assistants access to AWS CloudWatch Logs for analysis, searching, and correlation.
aws_resource_validator - AWS Resource Validator is a Python package that creates objects to validate, show constraints of common AWS resource names, and generate compatible patterns for tests. This helps ensure that AWS resource names comply with AWS naming rules and can be used for testing and validation purposes.
Artillery v2.0.23 - More regions for AWS Fargate tests, Improvements to the Slack integration, and Various fixes for TypeScript support, Playwright integration, and Azure ACI & AWS Fargate.
AWS Fundamentals - AWS Service Animations - Visual explanations of how AWS services and features work.
EnvGuard - Robust Environment Variable Validation for Modern Applications.
💡 DynamoDB Tip of the Week
Each week we have a quick DynamoDB tip from our resident DynamoDB expert, Uriel Bitton.
💡 “Use sort key to filter instead of filter expressions”
“Always try to design your sort keys to support filtering at the key level. It’s more efficient and avoids reading unnecessary items compared to using FilterExpression.”
Further reading: https://www.linkedin.com/pulse/how-perform-powerful-filtering-dynamodb-using-sort-key-uriel-bitton-iwcpe/
✖️ Social of the Week
This week’s social is on X/Twitter by “Uncle Bob“ Martin, on one of his new early morning “bathrobe rants”, this time on code reviews!
These have been really fun to watch, interesting, include satire (I hope), and all fundamentally have a lot of truth in them. Not many teams I have worked with use printed emails with red and yellow markers and projectors for code reviews (yikes!), but I do agree that pair and mob programming is often a much better approach to code reviews than single-person work.
Why? The collaborative process means that you are essentially reviewing the code, architecture, and approach together as you go, and also the benefits of removing the async nature (more time efficient), there are no single points of failure in knowledge, you fill in for each others weaknesses (perhaps somebody knows more about a service or framework than the other), and it is a great way to up-skill and increase knowledge within a team.
What are your thoughts on pair and mob programming compared to engineers working on tickets on their own? Feel free to leave a comment below.
🎙️ YouTube & Podcasts
Here are some of my favourite videos and podcasts this week.
⭐ My favourite video this week is by Mike Chambers and friends talking about Model Driven Agents using Strands. This is very cool!
Emrah Samdan and Allen Helton cover “Pushing Messages (and Limits) with Momento“.
Emily Bache covers “I Don't Need Another Scrum Master, Get Me a Technical Coach!“.
Mike Chambers does a great video titled “Model Driven Agents - Strands Agents (A New Open Source, Model First, Framework for Agents)“.
Patrick Kua discusses “Level Up: Choosing The Technical Leadership Path“.
Dave Farley covers “3 Reasons Your CI/CD Pipeline Isn't Working As It Should...“
The Primeagen has a great video on “How BAD Is Test Driven Development?“.
Weekly Case Study 🔍
This week’s case study comes from dacadoo where they moved from virtual machine to Kubernetes to finally serverless!
Dacadoo, a Swiss digital health company, reduced its cloud infrastructure costs by 78% and nearly eliminated operational overhead by evolving from an on-premises virtual machine setup to Kubernetes and finally to a fully serverless architecture on AWS.
Initially constrained by manual deployments and limited scalability, they first adopted Kubernetes to improve availability and global reach but found it complex to manage.
Transitioning to serverless using AWS services like Lambda, API Gateway, DynamoDB, and S3 allowed them to automate operations, scale effortlessly, meet regional compliance, and cut infrastructure maintenance to under an hour per year, all while improving performance and reliability.
🗣️ Inspirational Quotes and Thoughts
This week’s inspirational quote comes from American writer and humorist Mark Twain:
“Good judgement is the result of experience and experience is the result of bad judgement”
- Mark Twain
This quote is a reminder that making mistakes is part of the growth process, especially in software development and architecture on AWS, where you gain experience through building, and not purely through certifications. It is not a paper exercise to learn what works and what doesn’t, and you can’t buy that experience. Go build!
What are your thoughts and experiences with this quote? Feel free to leave a comment below.
🗳️ Poll of the Week
In last week’s poll, we asked the question “Are you using Cloudflare’s compute, database, storage, media and AI services?”.
Interestingly, 40% said “yes, some of them”, 20% said “other”, and 40% said no.
In my opinion, I am really excited about their service offerings, and I am excited to play with them more!
This week, we ask the question, “Do you version your microservices on AWS explicitly?”. When I say version, I mean through either tags, manifest files, package.json updates etc
Feel free to leave a comment below on why you chose your answer and your experiences!
📅 Serverless Events
The following serverless events are upcoming, so mark your calendars.
🎟️ To note, CFP is currently open for the AWS North Community Conference, which I am helping organise, and we also have opportunities for lightning talks throughout the day. Go check it out!
Other fantastic events happening soon:
ACD Switzerland - 22nd May 2025
ACD Bengaluru - 23rd May 2025
ACD Midwest - 5th June 2025
ACD Australia - 15th August 2025
ACD Adria - 5th Sept 2025
AWS Community Day Baltic - 10th Sept 2025
ACD Aotearoa - 18th Sept 2025
ACD Poland - 18th Sept 2025
ACD Portugal - 27th Sept 2025
ACD DACH - 7th Oct 2025
AWS North Community Conference - 16th Oct 2025
Do you have any upcoming events that you want to highlight? Message me below!
⭐ Spotlight
This week’s spotlight falls on Senior Solutions Architect on Powertools for AWS, Andrea Amorosi!
At AWS, Andrea plays a key role in the Powertools for AWS Lambda initiative (which many of us use every day in our serverless solutions - I know I do!), where he leads the TypeScript chapter. A passionate builder and prolific content creator, he has co-authored and edited numerous blog posts on the AWS Blog, as well as being a great international speaker!
Thank you for all you do for our wonderful Serverless community Andrea!
Thank you for reading the latest Serverless Advocate Newsletter!
If you want to find out a little more about me, please have a look at:
https://www.serverlessadvocate.com/
See you next time,
Lee
