Architecture Weekly

DNS Cache Memory Optimization 👷‍♂️ Whoever tells you data structures don’t matter in the age of AI - ban them. Cloudflare manages to save hundreds of terabytes of memory across the glove doing small optimizations to the ways DNS cache records are stored with the knowledge of how Rust store data and dropping dns owners when they are identical. Brilliant engineering work. #dns #performance

Efficient Software Factory at Uber 👷‍♂️ While everyone talks about making software factories, Uber actually does it. They dropped a banger of an article explaining the 4 levels of AI adoption in the company, cost structure, measurement approach and of course the results. Great job! #ai #casestudy

We started as pure Serverless. Here's what changed 2 years later https://softwarearchitectureweekly.substack.com/p/serverless-got-us-to-50-clients-then?utm_source=share&utm_medium=android&r=1m9i62

MicroVMs for throw-away jobs MicroVMs is a way to provide you a short-lived isolated execution environment. What are they best for? Right, security related tasks! How about running the virus scanning on docker images? While I am personally skeptical on mere signature-check scans in general and ClamAV in particular, the idea of running high risk payloads in isolated envs is really appealling. Checkout how to leverage MicroVMs for it. #security #aws

AWS EC2 Application Status Check 👷‍♂️ After decades of custom monitoring solutions, AWS introduced the every minute status check with HTTP probes. Auto Scaling groups can replace unhealthy instances based on this application status. Try it out and tell me in the comments how it improved your life! #ec2 #observability

AWS EC2 Application Status Check 👷‍♂️ After decades of custom monitoring solutions, AWS introduced the every minute status check with HTTP probes. Auto Scaling groups can replace unhealthy instances based on this application status. Try it out and tell me in the comments how it improved your life! #ec2 #observability

Should You Split Into Microservices? 👷‍♂️ We recently merged all our microservices(we had a handful) in a monolith. If you consider moving in a back direction, you need to ask yourself at least 5 questions on dependencies, different NFRs for the system parts, teams blocking each other, data boundaries and independent failures. Two candidate services, each owning its database, with red cross-boundary queries between them: one service, cut in half #microservices

Running a self-hosted LLM in Kubernetes with vLLM 👨‍💼 With the rise of cost for the LLMs and the privacy concerns more and more enterprises opt to run local models(and I am experimenting with them myself). Grab a guy how to setup an open-source LLM with Kubernetes! #llm #cloud #devops #architecture

We recently obtained SOC2 certification for Supplied. Our customers frequently ask how secure their data is with us. Answering this very question in the detailed post https://open.substack.com/pub/softwarearchitectureweekly/p/how-to-achieve-iso27001-and-soc2?r=1m9i62&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true #security #soc2

We became so faster writing code, but do we ship more? Talking with Baruch Sadogursky about what exactly prevents us from unlocking true productivity, and it's not better agents. 👇 https://youtu.be/a_Kq18ufZzU

Clustering Billions of Products for Agentic Commerce with Catalog API 🤓 Shopify Catalog groups billions of listings without a common schema. It first matches products inside each store, then connects them across stores with a Universal Product Identifier (UPI). LLMs assign a structured label to every product. This enables consistent grouping, high precision, and better recall. AI searches based on Catalog data convert twice as often as searches based on scraped data. #ai #architecture #softwareengineering #engineering

The State of Streaming to Apache Iceberg in July 2026: Every Path, Its Latency, and What to Do When Seconds Are Not Fast Enough 👨‍💼 In mid-2026, teams no longer ask, “Should we use Iceberg?” They ask, “How current can our Iceberg tables be?”. And this is where the main tradeoff relies dictating your data architecture and tools to go with. From tuned Flink to Kafka connect latency numbers varies from 30 seconds to 15 minutes. How to choose? Well, Alex Merced explains in his piece. #db #distributed #architecture #softwareengineering

Most “AI agents” are workflows with an LLM inside. The real difference: who controls the flow? In my new video, I break down the five parts of a real agent—prompt, tools, state, memory and loop—plus the production essentials: tracing, guardrails and evals. Here's the link: https://youtu.be/SmSv_6bI5QM

I bought a setup for running local LLMs. Grab the unpacking video! https://www.youtube.com/shorts/xcbmp1p06jM

People go to the technical conferences and the only value they get are free snacks and some talks missing the true purpose of such events. I published a guide how to actually prepare the conferences and what to do there depending on your career aspirations. https://open.substack.com/pub/softwarearchitectureweekly/p/capturing-value-out-of-technical?r=1m9i62&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true

Stragglers, Not Failures: How Adaptive Hedged Requests Reduce p99 Latency by 74 Percent 🤓 In fan-out microservice architectures, the dominant cause of high p99 latency is stragglers — slow-completing requests rather than failures — because one straggler in a fan-out blocks the entire composite response. While retries are a solution for failed requests, the stragglers require a parallel request if slow response is detected. Indeed new issues come with the solution, like write amplification, but if the goal is p99 optimized, raced requests are a way to go. The result is a 74% reduction in p99 latency with zero call-site configuration changes, with a reference implementation available as an open-source Go library. #distributed #architecture #engineering #softwareengineering

The Inference Paradox: How Split-Brain LLMs Are Killing Your GPU ROI 🤓 LLM inference has a structural hardware mismatch: the prefill phase is compute-bound (processing all input tokens in a single forward pass) while decode is memory-bandwidth-bound (reading the full KV cache to emit one token per step), so coupling both phases on the same GPU means each permanently starves the other. Kubex's enterprise audits surface average GPU utilization near 5% — monolithic vLLM serializes all prefill before decode can continue, and under high concurrency the delay compounds across every request in the batch. Disaggregating prefill and decode onto separate hardware-optimized node pools — as llm-d (now a CNCF sandbox project) implements on Kubernetes with prefix-cache-aware routing — yields 2–3x throughput at high concurrency by keeping the decode pool continuously active while the prefill pool handles bursts in parallel. #ai #llm #engineering #cloud

ScyllaDB Clusters at Discord, Zero Trust for AI Agents, CloudFlare Data Platform and many more in Architecture Weekly #198 https://www.youtube.com/watch?v=71AXRNKlg5c

Zero Trust for AI Agents 🍼 Each new technology bears security risks and the agentic systems is no exception. Both common vulnerabilities like supply chain attacks and excessive permissions stay; new ones like direct and indirect prompt injection, RAG poisoning and others comes. Follow the Antropic guide on applying Zero Trust to your agentic systems. #security #ai