Cvetkovic Milos

#JavaScript tech lead #AWS #SoftwareArchitecture

[Zach Lloyd @zachlloydtweets]

One interesting tradeoff as we've designed our factory infra is how much to support "if this, then that" determinism vs. using more free-form orchestration to move work through a factory.

We ended up making it flexible, where users can define the shape of the workflow they want through a "blueprint" that looks a bit like a DAG, but ultimately it's an orchestration agent that decides how to progress work, when to loop in a human, etc.

By orchestrator, I mean there isn't deterministic pattern matching like in Zapier to decide what happens next. Instead, an agent looks at the context and the guidance from the code blueprint and makes a decision, using something like a Jev classifier. And in situations where the guidelines are incomplete or a human asks for a different path, the agent is able to take it.

Curious how others are thinking about this.

[Zach Lloyd @zachlloydtweets]

Lots of discussion about what a "software factory" is.

Also useful to call out what a software factory is not:
1/ A loose collection of automations around cloud agents
2/ A team of "AI Engineers" that developers manage interactively
3/ A set of skills and MCPs that individual engineers all adopt across a team
4/ A personal agent swarm

These are all useful things, but they aren't the right primitives for engineering scalable agent systems across an organization.

The right primitives allow you to define all your agentic workflows in code, run them entirely in the cloud, and, crucially, measure, test and improve those workflows over time to maximize agent productivity while minimizing cost.

If you start thinking from first principles "what is the system we should build as a company to get the most out of coding agents" you will end up with something like this.

[SaltyAom @saltyAom]

I'm in a desperate situation right now that I can't manage all of the security advisories manually on my own especially when

1. I'm the only full-time core maintainer with community help
2. Have to put a lot of time and effort into Elysia 2.0 development that have been going on for 11 months now

I'm in a very uncomfortable situation right now that I don't have enough time/effort to maintain security high standard while developing the new major version and need additional tool to help me assist to maintain a high security standard

You have already seen what rogue AI and frontier AI could do and I'm not happy especially without somewhat near equivalent tool that can protect my users

I have been rejected with the identity verification for 3 time straight without any explicit reason and have been ghosted by support for 2 weeks now

If anyone know someone in Anthropic that can help with my situation, please I'm begging you, the situation is very dire right now

[Anthropic @AnthropicAI]

New on our Engineering blog: Creating in-house source control.

In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale.

Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: https://t.co/wzNkwHs7IN

[ClaudeDevs @ClaudeDevs]

New blog: Building agents that reach production systems with MCP.

When should agents use direct APIs vs CLIs vs MCP? Plus patterns for building MCP servers, context-efficient clients and pairing MCP with skills.

https://t.co/Q4UrUVgVYB

[DHH @dhh]

In 2023, we spent $3,934,099 on AWS + other hosting. In 2026, our hosting + support bill is down to ~$1m/year due to the cloud exit. Even including all the hardware buying, we will already have saved ~$4m by the end of this year. And going forward, it's ~$3m/yr in savings 🤑

[Addy Osmani @addyosmani]

Want to give your agent quality checks?

Chrome's DevTools MCP now includes:

⚡️ Performance checks via Lighthouse
📈 Memory leak detection Skill
🦻 Accessibility debugging Skill
🎨 LCP optimization Skill

and an experimental new CLI 👀

[Matteo Collina @matteocollina]

Why Node.js Is the Critical Enabler for AI Applications https://t.co/a17JwPLihK

[Uncle Bob Martin @unclebobmartin]

Morning bath robe rant: it’s a good thing.

[Daily Dose of Data Science @DailyDoseOfDS_]

Learn how LLMs work under the hood!

This is the best place to visually understand the internal workings of a transformer-based LLM.

Explore tokenization, self-attention, and more in an interactive way:

[Anton Zhiyanov @ohmypy]

Here is how to do SQL upserts in all major databases.

With interactive examples so you can read and practice.

https://t.co/fKXukYWyny

[Filip Pyrek @FilipPyrek]

How wrong is this slide in my presentation?

[Cvetkovic Milos @milos_dev]

@juristr @NxDevTools Clear, only on AWS :)

[Nx @NxDevTools]

You could always use Nx also for just single app development (rather than a monorepo). How? By leveraging the powerful generators & maybe even using libs for better structuring/modularization of the app.

What if we told you that it might become even easier very soon 🤔

[Simon @simonholdorf]

Want to become an AWS professional and snag a high-paying (remote) cloud job?

Follow & build these 9 projects to acquire real-life working experience for your resume:

[Alex Brogan @alexbrogan]

Most people suck at decision-making.

Let's change that.

Here are 15 of the best decision-making mental models I've found:

[Simon @simonholdorf]

If you want to become better at DevOps & SRE, you have to practice, practice, practice 🛠️

I found a repository containing exercises and questions, including answers to help you understand DevOps better and even prepare for interviews. 💪

You can find the link here 👇

[Towards AWS @TowardsAws]

Over 160,000+ jobs on LinkedIn require AWS (the USA alone). 🤯

It is one of the most demanding skills in the industry today!

Best resources to learn:🧵

[Cvetkovic Milos @milos_dev]

@MarkoStDev @angular @DanWahlin @SantoshYadavDev @twerske @ngrx_io @GoogleDevExpert Congrats, Marko, well deserved :)🍻