Zach S.

Just tried the new Claude Agent SDK from Anthropic and it’s actually very good. Interesting part is its built-in session and memory management. The SDK explicitly supports automatic context-management and session handling so your agent doesn’t run out of context. The companion feature from Anthropic, the “memory tool” enables the model to read and write persistent files (in a /memories directory) between sessions, allowing it to maintain project context over time. Real session state, real memory, real continuity. Exciting times.

73

1 Comment

Ever wonder how far you can actually scale n8n? I’ve run it on everything from a $6 DigitalOcean droplet to a Redis-powered queue mode setup handling over 400,000 executions a month. Yesterday, Mike Hostetler asked me: “Do you host n8n yourself, or do you use their cloud offering? What would you do now if you had to set it up again?” So I wrote up everything I’ve learned—what works, when to upgrade, and what breaks as you grow: ☁️ Beginner: Start with n8n Cloud or a $6 SQLite droplet—super fast to get going. 🐘 Advanced: Upgrade to PostgreSQL (we use Supabase) when you hit DB or team-size limits. 🚀 Scale: Use queue mode with Redis when you need serious concurrency and performance. Full breakdown (with upgrade signals, pricing, and setup tips) is here: 👉 https://lnkd.in/gsEdqbSd (Some links in the post are affiliate links—if you use them, it helps support more guides like this!)

81

31 Comments

There's been some talk around storage unification and zero-copy lakehouse integrations recently, and I wanted to better define these terms as well as look at ways we should evaluate different design decisions in this space. So I’ve published a new post: A Conceptual Model for Storage Unification Storage unification isn't about “one storage format to rule them all,” but virtualizing them into a unified logical model. In this post, I lay out a conceptual framework for thinking about storage unification (aka data virtualization). I explore: 🔹 Internal vs. shared tiering (and why shared tiering is so tricky). 🔹 When materialization makes more sense. 🔹 Client vs. server-side stitching. 🔹 Lifecycle and schema management across evolving formats. 🔹 Seven key considerations for evaluating different approaches. This is the groundwork for a set of posts where I’ll dig into how some real-world systems deliver storage unification, primarily about unifying real-time and historical data. https://lnkd.in/dBiS2kxJ

45

Early in my career everything was about building large scale homogeneous systems. What’s been more interesting over the last 5y has been heterogeneous multi cloud systems. The tradeoffs are more complex to build simple (irreducible complexity) and elegant systems in that space.

41

2 Comments

Today's outages across both GCP and AWS are a stark reminder of a simple truth: your reliability is only as resilient as the platforms you build on. It's easy to treat cloud providers as "infrastructure solved," but incidents like today expose just how quickly cascading failures can happen — especially when your incident tooling is also downstream of the same providers you're trying to monitor or manage. We saw several mission critical incident management platforms go down today during the exact moment customers needed them most. And to be clear: this isn't about dunking on anyone. Infra is hard. Outages happen. We've been there. But this is exactly why Rootly was built with a multi-cloud, fault-isolated architecture (we are the only ones). We run across providers, we separate control and data planes, and we assume failure at every level of the stack — including ours. Because when you're on-call, "it's down" isn't an acceptable answer from the tool meant to help you respond. Reliability of your incident management platform is the thing I talk most about to our enterprise customers and we are seeing that pay off today. Enterprise customers consistently hear me highlight our platform’s reliability and today that focus is paying dividends. If today felt stressful, you're not alone. But if it makes you revisit your own assumptions around redundancy, failure domains, and operational independence — that's a good thing.

100

3 Comments

Artillery Cloud is now powered by PlanetScale Metal. I'm a long time fan of PlanetScale's so as soon as they launched GA of their new Postgres product we started looking at switching. Very smooth process overall, the docs and DX is excellent and miles ahead of AWS. If you're a user you should notice the app feeling quite a bit faster. Otherwise it's business as usual. For us, the built in query analyzer has already highlighted a couple of improvements to make, so the app will get even faster again in the upcoming few days!

35

2 Comments

Today at #MSBuild conference we are releasing following updates to Teams AI Library to build collab agents: ➡️ A2A - enable peer to peer communication between agents https://lnkd.in/gUYzu8Fc ➡️ Memory - enable agents in Teams to remember past user interactions, enabling more personalized and context-aware experiences https://lnkd.in/g_eGpM7h ➡️ Simplified DevEx - With just few lines of code you can have your first agent with MCP/A2A working in Teams To learn more about Teams AI Library check it out here: aka.ms/teamsaiv2

109

3 Comments