Latest posts
Architecture Weekly
17 Sept, 07:01
Uber Eats Search Pipeline 👷♂️ Search Performance is always - and I mean it - always is a holistic problem. The engineers need to search through UX approaches, caching, retrieval mechanics, query performance and of course how the full stack operates as a whole. Uber Eats shows how they touch upon every aspect, including running an AI agent with the live data profiler to fix the issues in search.#performance #casestudyUberHalving the Time: How Uber Eats Rebuilt Its Search PipelineEvery millisecond of latency is a lost opportunity—especially when people are hungry. 🍔 We cut Uber Eats search latency in half by re-architecting our full-stack pipeline. Read our latest deep dive to see how we did it.Architecture Weekly
16 Sept, 07:56
Github 5 Incidents in August 👷♂️ The Github availability became a living joke this year, but not because of the engineering capability, but because of sheer amount of pull requests the ai agents now generate. Github shares their 5 August incidents, the reasons behind them and what they learned from them. First hand experience from the biggest code storage in the world!#ai #reliabilityThe GitHub BlogGitHub availability report: August 2026In August, we experienced five incidents that resulted in degraded performance across GitHub services.Architecture Weekly
9 Sept, 07:28
Local models will not win 🍼 I tried to see if local models are viable for serious development. Spoiler: they are not. Sean Goedecke agrees: everybody tends to choose the most powerful model in their price range, and only datacenters have the corresponding capacity. Plus, they are well optimized for it. So local models will just not happen, forget it.#aiSeangoedeckeNo, local models will not winEvery time a new open-weight AI model is released, people say that local models are the future. Why spend billions of dollars building out datacenters when…Architecture Weekly
5 Sept, 07:58
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 #performanceCloudflare BlogHow we saved 100 terabytes of memory by optimizing 1.1.1.1’s DNS cacheFive Rust-level memory optimizations to the DNS cache layout of Big Pineapple cut per-entry memory by 56%, freeing approximately 100 TB of memory across Cloudflare's fleet.Architecture Weekly
31 Aug, 10:42
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 #casestudy1,410Open in TelegramArchitecture Weekly
17 Aug, 08:12
We started as pure Serverless. Here's what changed 2 years laterhttps://softwarearchitectureweekly.substack.com/p/serverless-got-us-to-50-clients-then?utm_source=share&utm_medium=android&r=1m9i62SubstackServerless Got Us to 50 Clients. Then It Became the Problem.From AWS Lambdas and DynamoDB to a single container in ECS backed by PostgreSQL.Architecture Weekly
14 Aug, 08:54
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 #awsPubudu.DevScan user uploaded files using AWS Lambda MicroVM and ClamAVAWS Lambda MicroVMs are isolated serverless compute environments that provide vm-level isolation. In this post, I discuss a scenario where user uploaded files can be scanned for malware using MicroVM and ClamAV an open-source antivirus engine.1,480Open in TelegramArchitecture Weekly
12 Aug, 12:58
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 #observabilityAmazonAmazon EC2 introduces application status checks - AWSDiscover more about what's new at AWS with Amazon EC2 introduces application status checksArchitecture Weekly
11 Aug, 15:13
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#microservicesSubstackSOC2 journey at Supplied, Interview with JBaruch, Splitting into Microservices and many more in Architecture Weekly #201.Also today: Optimizing Prompt Cache, AWS EC2 Application Status Check , MicroVMs for throw-away jobs, Writing code was never the point and Meerkat: Cloudflare Consensus Solution.Architecture Weekly
5 Aug, 10:42
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 #architectureCNCFRunning a self-hosted LLM in Kubernetes with vLLMRunning large language model (LLM) workloads in-house is one of several patterns teams adopt alongside managed API services. Managed API services are convenient and well suited to many workloads.Architecture Weekly
4 Aug, 11:25edited
SOC2 demystifiedWe 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#security #soc2SubstackHow to achieve ISO27001 & SOC2 as a B2B startupCertification process, technical measures and policies to get your SOC2 stamp.Architecture Weekly
30 Jul, 07:04
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_Kq18ufZzUYouTubeAI Agents Made Developers Faster. Why Aren’t Teams Shipping More? | Baruch SadogurskyAI agents have transformed how quickly developers can produce code. But organizational productivity hasn’t increased at the same speed. In this Architecture Weekly conversation, Baruch Sadogursky and Vladimir Ivanov examine what actually changed in AI-assisted…1,490Open in TelegramArchitecture Weekly
22 Jul, 06:05
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
1,610Open in TelegramArchitecture Weekly
21 Jul, 16:38
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 #softwareengineeringSubstackThe State of Streaming to Apache Iceberg in July 2026: Every Path, Its Latency, and What to Do When Seconds Are Not Fast EnoughThe most common architecture question I get in 2026 is no longer “should we use Iceberg.” That one is settled.Architecture Weekly pinned «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.…»
17 Jul, 15:59
Architecture Weekly
17 Jul, 14:33
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_6bI5QMArchitecture Weekly
14 Jul, 12:39
I bought a setup for running local LLMs. Grab the unpacking video!https://www.youtube.com/shorts/xcbmp1p06jMYouTubeUnboxing an LLM BeastUnboxing a new workstation I plan to use for local LLMs like Qwen3....Architecture Weekly
2 Jul, 05:57
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=trueSubstackCapturing Value Out of Technical ConferencesAnd it's far from listening to talks.Architecture Weekly
29 Jun, 06:46
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 #softwareengineeringInfoQStragglers, Not Failures: How Adaptive Hedged Requests Reduce p99 Latency by 74 PercentIn fan-out microservice architectures, slow-but-completing requests accumulate across services and drive p99 latency far higher than per-service metrics suggest. This article presents an adaptive hedging mechanism that uses DDSketch for real-time quantile…Architecture Weekly
24 Jun, 09:37
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 poolKubexThe Inference Paradox: How Split-Brain LLMs Are Killing Your GPU ROIWe take a deep dive into the intricacies of LLM inference to expose why low GPU utilization is a fundamental architectural issue, not a hardware deficiency.
Related Channels
Other channels in the same section of the catalogue.
