# Experiment Published articles for Experiment. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## One or two nameservers? DevFeed: [One or two nameservers?]() Original publisher: [Read original article]() Author: Geoff Huston Published: 2026-09-15T06:01:49Z Content type: article Language: en Sources: [APNIC Blog]() Topics: [Internet](), [Query (disambiguation)](), [Server](), [Caching]() Tags: [2026](), [africa](), [article](), [china](), [dns](), [europe](), [experiment](), [india](), [internet](), [ipv4](), [ipv6](), [measurement](), [measurements](), [recursive-resolver](), [repeat](), [responses](), [server](), [tech-matters]() ### AI overview This article reports an experiment testing whether serving a DNS zone with two authoritative dual-stack nameservers changes repeated queries. Compared with one nameserver, two nameservers increased the single-query completion rate from 58% to 71% and reduced the average queries per test from 3.43 to 2.57. ### Source excerpt Do multiple dual-stack nameservers increase or decrease repeat DNS queries? The results of this experiment were a complete surprise. ## An experiment with Gemma 4 on a Raspberry Pi explores what an LLM would do with its own computer DevFeed: [An experiment with Gemma 4 on a Raspberry Pi explores what an LLM would do with its own computer]() Original publisher: [Read original article]() Author: Alex Ewerlöf Published: 2026-09-14T09:09:42Z Content type: opinion Language: en Sources: [Alex Ewerlof Notes]() Topics: [gemma4](), [Large Language Model](), [Raspberry Pi](), [Go Language](), [Simulation]() Tags: [computer](), [data](), [experiment](), [gemma-4](), [go](), [llm](), [patterns](), [raspberry-pi]() ### AI overview The author describes an experiment in which Gemma 4 12B was asked what it would do with its own computer. Running on a Raspberry Pi 1 through a Go harness, the model imagined ingesting large amounts of data, finding patterns, and running large-scale simulations. ### Source excerpt LLM's response ## Turning Names Into Digital Architecture with Three.js DevFeed: [Turning Names Into Digital Architecture with Three.js]() Original publisher: [Read original article]() Author: Serhii Polyvanyi Published: 2026-09-09T17:55:00Z Content type: article Language: en Sources: [Codrops]() Topics: [Three.js community](), [Three.js custom shaders](), [Three.js shaders]() Tags: [3d](), [architecture](), [articles](), [best-free-frontend-development-resources](), [best-javascript-libraries-for-animations](), [best-web-development-blogs](), [conference](), [creative-coding](), [creative-ui-ideas-for-web-design](), [css-animations](), [css-grid-vs-flexbox](), [css-hover-effects](), [css-only-effects](), [experiment](), [frontend-development](), [gsap-animation-examples](), [how-to-build-a-responsive-website-from-scratch](), [how-to-create-parallax-scrolling-effects](), [how-to-create-smooth-css-transitions](), [how-to-make-a-portfolio-website-interactive](), [how-to-optimize-animations-for-performance](), [html5-and-css3-tricks](), [javascript-effects](), [javascript-ui-interactions](), [micro-interactions-for-websites](), [modern-web-development](), [react-ui-animations](), [step-by-step-guide-to-webgl-animations](), [svg-animations-in-css](), [three-js](), [three-js-3d-animations](), [three-js-conference](), [ui-ux-design-tips](), [web-design-tutorials](), [webgl-tutorials-for-beginners](), [wordpress-themes-and-plugins]() ### AI overview An interactive Three.js experiment turns typed names into lightweight, metallic-looking 3D letter structures. It explains contour generation, interpolation and deformation, structural line geometry, and a custom shader with a baked reflection map. ### Source excerpt For the Three.js Conference in Paris, BL/S® created an interactive experiment that transforms your name into a dynamic 3D structure. ## A switchback career move from engineering management to software engineering DevFeed: [A switchback career move from engineering management to software engineering]() Original publisher: [Read original article]() Author: Anton Zaides Published: 2026-09-08T06:01:00Z Content type: opinion Language: en Sources: [Manager.dev]() Topics: [Job](), [AI-assisted coding](), [Software](), [Tech Lead]() Tags: [agents](), [career](), [data](), [engineering-manager](), [experiment](), [software-engineer]() ### AI overview The author describes moving from engineering management back to software engineering after struggling to secure another Engineering Manager role. The article also discusses an experiment in which an engineering team worked more deliberately with coding agents and a tech lead. ### Source excerpt The perfect time for a switchback career move ## Visualize how CUPED adjusts experiment results with Datadog DevFeed: [Visualize how CUPED adjusts experiment results with Datadog]() Original publisher: [Read original article]() Author: Tyler Buffington; Lukas Goetz-Weiss; Ryan Lucht Published: 2026-09-01T00:00:00Z Content type: tutorial Language: en Sources: [Datadog | The Monitor blog]() Topics: [experiments]() Tags: [analysis](), [digital-experience-monitoring](), [experiment](), [experiments](), [how-to](), [metrics](), [product-analytics](), [visualization]() ### AI overview This tutorial explains Datadog Experiments' CUPED adjustments visualization, which breaks the difference between raw and CUPED-adjusted experiment lift into individual covariate contributions. ### Source excerpt Learn how Datadog visualizes CUPED adjustments so you can trace which covariates change experiment lift estimates and improve precision. ## Better answers, broader thinking: What students gain from ChatGPT and critical-thinking training DevFeed: [Better answers, broader thinking: What students gain from ChatGPT and critical-thinking training]() Original publisher: [Read original article]() Published: 2026-08-27T09:00:00Z Content type: article Language: en Sources: [OpenAI News]() Topics: [ChatGPT](), [Artificial Intelligence](), [OpenAI]() Tags: [ai](), [analysis](), [business](), [chatgpt](), [company](), [experiment](), [generate](), [gpt](), [marketing](), [openai](), [performance](), [reasoning](), [research](), [training]() ### AI overview Un randomized experiment with more than 1,000 first-year undergraduate students found that ChatGPT access improved the quality, coherence, logic, and professionalism of work on a real-world business assignment. Separate causal-reasoning training increased the originality and variety of ideas, and students receiving both interventions showed both benefits. ### Source excerpt A randomized study of more than 1,000 students examines ChatGPT, critical thinking, originality, and student performance on a real-world university assignment. ## When Can LLMs Replace Humans in A/B Tests? DevFeed: [When Can LLMs Replace Humans in A/B Tests?]() Original publisher: [Read original article]() Author: Spotify Engineering Published: 2026-08-13T18:57:22Z Content type: article Language: en Sources: [Spotify Engineering](), [Spotify Engineering Blog]() Topics: [Large Language Model](), [A/B Testing](), [datasets](), [data](), [experiments]() Tags: [a-b-testing](), [analysis](), [data](), [data-science](), [experiment](), [gpt](), [llm](), [llms](), [model](), [testing]() ### AI overview The article examines whether large language model predictions can replace human outcomes in A/B tests. Using the Upworthy dataset, it finds that calibrated predictions can recover treatment effects under specific assumptions, while raw predictions recovered only 39% of the observed human effect. ### Source excerpt TL;DR: LLM predictions can stand in for human outcomes in A/B tests, but only by assumption, not by design.... The post When Can LLMs Replace Humans in A/B Tests? appeared first on Spotify Engineering. ## Introducing Mintlify Index DevFeed: [Introducing Mintlify Index]() Original publisher: [Read original article]() Author: Kyan Yang Published: 2026-08-06T00:00:00Z Content type: release Language: en Sources: [Mintlify Blog]() Topics: [Documentation](), [AI-assisted coding](), [API](), [MCP](), [MCP Server]() Tags: [announcements](), [api](), [beta](), [coding-agents](), [documentation](), [experiment](), [mcp](), [mcp-server](), [retrieval]() ### AI overview Mintlify introduces Mintlify Index, a retrieval layer that aggregates documentation from 9,000 Mintlify-powered sites for coding agents. It routes documentation-related queries to optimized retrieval and other queries to an external web search provider, with fallback between sources. The article reports an experiment in which Index was preferred in 96 of 150 implementation-planning tasks and completed tasks 48% faster end-to-end, with higher factual correctness and citation support and fewer critical errors. Mintlify launched Index as a free beta with an API and MCP server. ### Source excerpt Mintlify Index aggregates the docs of 9000 Mintlify-powered documentation sites into a natural language retrieval layer for coding agents. ## Explorers, exploiters, and the myth of the 100x engineer DevFeed: [Explorers, exploiters, and the myth of the 100x engineer]() Original publisher: [Read original article]() Author: Eira May Published: 2026-08-05T07:40:00Z Content type: article Language: en Sources: [Stack Overflow Blog]() Topics: [Artificial Intelligence](), [AI Strategy](), [Reinforcement learning](), [leaders-of-code]() Tags: [ai](), [ai-adoption](), [ai-tools](), [business](), [engineering](), [engineering-leadership](), [experiment](), [explore](), [leaders-of-code](), [learning](), [podcast](), [productivity]() ### AI overview The article challenges the myth that a small group of inherently exceptional engineers drives disproportionate AI gains. It argues that engineering leaders should treat exploration and exploitation as a continuum and help more engineers develop through curiosity, adaptability, experimentation, and learning, rather than focusing only on identifying and promoting existing "100x" performers. ### Source excerpt The "find the special ones and promote their traits" approach isn't the best or only way to drive AI adoption and productivity on an engineering team. ## Research: How we cut AI costs by 80% DevFeed: [Research: How we cut AI costs by 80%]() Original publisher: [Read original article]() Author: Zohar Einy Published: 2026-07-30T11:24:23Z Content type: article Language: en Sources: [Developer Experience & Platform Engineering Blog | Port]() Topics: [Artificial Intelligence](), [context window](), [MCP](), [semantic-layer](), [Claude](), [data](), [sdlc](), [GitHub](), [Caching]() Tags: [agents](), [ai](), [claude](), [context-window](), [cost](), [data](), [experiment](), [github](), [mcp](), [reasoning](), [sdlc](), [semantic-layer](), [tokens]() ### AI overview This research article examines rising AI costs caused by messy, repeatedly assembled context for agent queries. An experiment using thousands of production queries and a test set of 1000 commonly asked SDLC queries found that structured context was 80% cheaper than unstructured context. The proposed approach pre-relates context and uses a semantic layer to reduce data hops, reasoning, and token consumption. ### Source excerpt We ran thousands of AI queries on unstructured and structured context and measured the cost. Structured context was 80% cheaper than unstructured. ## Agent platform (Part 1): How we help Grab build and run AI agents at scale DevFeed: [Agent platform (Part 1): How we help Grab build and run AI agents at scale]() Original publisher: [Read original article]() Author: Kendrick Tan Published: 2026-07-24T00:00:10Z Content type: article Language: en Sources: [Grab Tech]() Topics: [AI Bots](), [AI Chat](), [MSP MCP]() Tags: [ai-agents](), [ai-infrastructure](), [architecture](), [design](), [engineering](), [experiment](), [generative-ai](), [go](), [llm](), [machine-learning](), [mcp](), [platform](), [scale]() ### AI overview Grab describes the first part of its AI-agent platform story: a Tech Infra support bot that evolved from operational automation into a GPT-powered Level-0 support layer. The article outlines its reasoning and tool planes and explains how scaling challenges informed the LLM-Kit framework. ### Source excerpt Part 1: From one support bot to a framework At Grab, AI agents have evolved from interesting team prototypes into production services used every day by millions of merchants, drivers, and consumers. Today, more than 500 services run on our internal agent framework, over 50 Model Context Protocol (MCP) servers are registered on our remote MCP framework, and a single Large Language Model (LLM) gateway fronts every model call across the company, handling billions of tokens each month. None of this was designed up front. It began as the plumbing behind one internal support bot, which then expanded because the same problems kept resurfacing for every team trying to ship an agent. This series tells the story of what the platform eventually became. This Part 1 of the blog focuses on the beginning: the architecture of our AI support bot, the specific pain points we hit while scaling and iterating on it, and how each of those failures became a core building block in the framework we now call LLM-Kit. The bot that started it Imagine you have a question for the Technical Infrastructure (Tech Infra) team - the engineers who run the cloud platforms, databases, developer tooling, and AI infrastructure behind Grab's ecosystem. Instead of immediately paging an on-call engineer, a bot first triages the request, checks the team's documentation, runbooks, and past Slack threads, and tries to answer directly in the thread. If it still cannot resolve the issue, it routes the ticket to the right human, with the relevant context already attached. That is what we built with the Tech Infra Support Bot. In the first half of 2023, Tech Infra handled thousands of support tickets, many of them repeated questions that had already been answered somewhere internally. Before LLMs, the bot's role was mainly operational; performing tasks like helping track acknowledgments and response times for on-call engineers. With the arrival of GPT-4-32k, we evolved it into a GPT-powered Level-0 support layer th ## Start Customizing NVIDIA Nemotron 3 Nano with Prime Intellect Lab in Minutes DevFeed: [Start Customizing NVIDIA Nemotron 3 Nano with Prime Intellect Lab in Minutes]() Original publisher: [Read original article]() Author: Chris Alexiuk Published: 2026-07-23T16:00:00Z Content type: tutorial Language: en Sources: [NVIDIA Developer](), [NVIDIA Technical Blog]() Topics: [rlvr](), [Large Language Model](), [Fine-tuning](), [Reinforcement learning](), [Python](), [coding](), [math]() Tags: [agentic-ai-generative-ai](), [blog](), [coding](), [customization](), [developer-tools-techniques](), [developers](), [experiment](), [featured](), [fine-tuning](), [getting-started](), [math](), [nemotron](), [nvidia](), [open](), [python](), [reinforcement-learning](), [rlvr](), [top-stories](), [train](), [training](), [tutorial]() ### AI overview This tutorial shows how to customize NVIDIA Nemotron 3 Nano with Prime Intellect Lab using reinforcement learning with verifiable rewards on a Python Math environment. It covers a baseline-training-reevaluation workflow and produces a downloadable LoRA adapter. ### Source excerpt Customization is what enables developers to take a general model and tailor it to use cases, domains, languages, and more. However, customization comes with a... ## Raspberry Pi Projects That Sound Great But Are Usually a Bad Idea DevFeed: [Raspberry Pi Projects That Sound Great But Are Usually a Bad Idea]() Original publisher: [Read original article]() Author: Patrick Fromaget Published: 2026-07-18T05:00:00Z Content type: opinion Language: en Sources: [RaspberryTips]() Topics: [Hardware](), [Cryptocurrency](), [hosting](), [Linux](), [Security](), [servers](), [1Password in the browser]() Tags: [crypto](), [experiment](), [hardware](), [hosting](), [inspiration](), [linux](), [projects](), [raspberry-pi](), [security](), [self-hosting]() ### AI overview The article argues that although Raspberry Pi computers can run many projects, some are poor choices for long-term reliance because they require more reliability and maintenance than a Pi setup can provide. It presents cryptocurrency mining and self-hosting an email server as examples: both can be useful learning exercises, but Raspberry Pi hardware is generally unsuitable for profitable mining or dependable email hosting. ### Source excerpt After years of testing and writing tutorials on RaspberryTips, I can tell you one thing: you can make your Raspberry Pi do almost anything. However, just because it works doesn't mean it's a good idea. Let's talk about it. A Raspberry Pi can run many home projects, but some are poor choices for long-term use.... ## Sightlines issue no.1: Insights from Config DevFeed: [Sightlines issue no.1: Insights from Config]() Original publisher: [Read original article]() Author: Figma Published: 2026-07-16T16:37:00Z Content type: article Language: en Sources: [Figma Blog]() Topics: [Artificial Intelligence](), [OpenAI]() Tags: [ai](), [beginner](), [collaboration](), [design](), [development](), [engineering](), [experiment](), [insights](), [leadership](), [product-development](), [prototyping]() ### AI overview Sightlines issue no.1 shares lessons from Config's Leadership Collective about leading product, design, and engineering teams in the age of AI. It emphasizes enduring leadership fundamentals, experimentation, collaboration, human judgment, and maintaining quality as teams adopt AI. ### Source excerpt The conversations we're still replaying from Config's Leadership Collective. ## How Prompt Tuning Improved GPT-5.5 in VS Code DevFeed: [How Prompt Tuning Improved GPT-5.5 in VS Code]() Original publisher: [Read original article]() Author: VS Code Team Published: 2026-07-06T00:00:00Z Content type: article Language: en Sources: [Visual Studio Code - Code Editing. Redefined.]() Topics: [AI Chat](), [AI Bots]() Tags: [agent](), [blog](), [coding](), [experiment](), [gpt](), [openai](), [testing](), [tools](), [validation](), [vs-code]() ### AI overview VS Code and OpenAI tested two system-prompt variants for GPT-5.5 coding agents to reduce exploration and validate edits sooner. The experiment compared the variants with a control on live traffic and shipped the stronger result. ### Source excerpt See how VS Code and OpenAI tested GPT-5.5 system prompt changes in a two-week experiment, cutting tool calls and tail-end token usage while speeding up edits. Read the full article ## 7 questions we had going into Config Leadership Collective DevFeed: [7 questions we had going into Config Leadership Collective]() Original publisher: [Read original article]() Author: Emma Webster Published: 2026-07-03T02:45:00Z Content type: article Language: en Sources: [Figma Blog]() Topics: [Artificial Intelligence](), [User experience (UX)](), [Figma]() Tags: [ai](), [design](), [experiment](), [leadership](), [learning](), [ux]() ### AI overview Leadership lessons from Config Leadership Collective: as AI and creative tools rapidly change design and product work, leaders are adapting their processes, experimenting hands-on, and giving teams permission to learn. The article also emphasizes enduring fundamentals such as understanding users, supporting their workflows, and maintaining craft and care. ### Source excerpt We showed up to Leadership Collective hoping to learn how leaders are guiding their teams through change, what expertise means now, and how they're keeping quality high as the pace picks up. Here's what we heard. ## The effect distribution: The missing piece in experimentation programs DevFeed: [The effect distribution: The missing piece in experimentation programs]() Original publisher: [Read original article]() Author: Tyler Buffington Published: 2026-07-02T00:00:00Z Content type: article Language: en Sources: [Datadog | The Monitor blog]() Topics: [experiments]() Tags: [a-b-testing](), [experiment](), [experimentation](), [experiments](), [testing]() ### AI overview The article explains why effect distributions are essential for interpreting results across experimentation programs. It shows how statistically significant results can all be false positives when true effects are concentrated at zero, and introduces the challenge of estimating true effects from noisy observed effects. ### Source excerpt Learn about the importance of considering the effect distribution when running experiments. ## Variance Reduction Below the Randomization Grain DevFeed: [Variance Reduction Below the Randomization Grain]() Original publisher: [Read original article]() Author: Tilman Drerup Published: 2026-07-01T16:28:36Z Content type: article Language: en Sources: [Instacart]() Topics: [experiments]() Tags: [capacity](), [causal-inference](), [economics](), [engineering](), [experiment](), [experimentation](), [experiments](), [machine-learning](), [marketplaces](), [metrics](), [reduce](), [science](), [techniques](), [variance]() ### AI overview This article explains how marketplace experiments can reduce metric variance below the level at which treatment is randomized. It describes cluster-level randomization for containing interference and shows how fine-grained outcome predictability can improve statistical power and reduce experimentation time. ### Source excerpt Sergio Camelo, Caitlin Kearns, Matias Cersosimo, and Tilman Drerup As artificial intelligence increases the velocity of engineering and science teams, experimental throughput is set to become a bottleneck for many product decisions. Many companies can now build faster than they can experiment, with queues of good ideas running the risk of not being tested because of lack of experimental capacity. This problem is particularly severe in marketplaces, where the presence of spillover and cannibalization effects between experimental units requires cluster-level randomization techniques. That randomization, in turn, has the unfortunate tendency to substantially reduce statistical power and slow down experimentation. In this post, we show that the predictability of outcomes at fine grains can be exploited to reduce the variance of aggregate metrics, even when experiments themselves are run at a coarse level. Since statistical power depends on metric variability, this yields considerable reductions in experimentation time. The Interference Problem In marketplace settings, behavior and outcomes for individual participants are inherently intertwined. In a delivery marketplace like Instacart, for example, the dispatch system solves a bipartite matching problem between shoppers and customer orders. Since assignments are global and interdependent, matching an order to one shopper means that the same order cannot be matched to another shopper. As a result, changing the handling for a single order creates ripples that affect the orders around it. If an experimenter were to assign a treatment intervention to one of these orders while leaving neighboring orders as controls, the latter would evidently be contaminated. A common response to this problem is to randomize treatments at the level of a cluster, chosen so that interference can stay within it. In food and grocery delivery, that cluster is typically a geographical region. Since every order within a region sees the same treatme ## How We Built an AI Agent to Clean Up Dead Code After A/B Tests DevFeed: [How We Built an AI Agent to Clean Up Dead Code After A/B Tests]() Original publisher: [Read original article]() Author: Aseem Upadhyay Published: 2026-06-23T10:54:43Z Content type: article Language: en Sources: [Housing.com]() Topics: [AI Agent](), [experiments](), [Code](), [Pull Request](), [context](), [Android]() Tags: [agent](), [ai](), [ai-agent](), [ai-agents-in-action](), [automated](), [code](), [concurrent](), [experiment](), [experiments](), [infrastructure](), [jira](), [llm]() ### AI overview Housing.com describes building an AI-agent pipeline to help clean up code after A/B tests. The workflow interprets experiment tickets, reports experiment status, and applies instructions to code, while using scripts for deterministic steps and an LLM where judgment is required. ### Source excerpt Photo by Microsoft Copilot on Unsplash At Housing.com, running product experiments is a continuous cycle. A/B tests go live, collect data, and eventually reach a conclusion. That's the exciting part. Then comes the mundane reality where someone has to clean up the code .i.e. remove a feature flag, promote a winning variant or revert the loser variant and finally raise a change request. Sounds simple? Maybe Is it tedious and quietly expensive? Yes! lifecycle of a taskThe Problem Worth Solving An experiment conclusion ticket typically lands on an engineer's desk looking something like this: Experiment: show_listing_map_widget Platform: Android Result: Negative - revert to control The job of the assigned engineer involves four distinct steps: Find every reference to the flag across the codebase. Delete the losing variant's code path. Trace every side-effect that only existed to support that variant Commit, open a PR, and comment on the Jira ticket. Step 3 is where the trap lies. Be it applying or removing a change, changing all the infrastructure code dependent on it could increase the complexity and risk of creating technical debt. But what if we automated a part of it? The AI Agent Pipelineupdated AI enabled lifecycle The problem statement became simple: Let stakeholders own the trigger. We built two agents to make it happen, 1. to interpret tickets and report experiment status 2. to take the instructions and code. Then came the hard part. Navigating Roadblocks The real complexity lies in building an AI agent that runs autonomously and serves different users across different use cases We found ourselves wrestling with questions we hadn't fully anticipated: How do we optimise on the tokens used per request? How do we handle concurrent requests? How do we ensure that the consistency in the output? So we went looking for answers.. Optimising Tokens per request Not every step needs AI. At each point in the workflow, we asked one question: is this operation deterministic ## How Evaluation-Driven Development (EDD) Works DevFeed: [How Evaluation-Driven Development (EDD) Works]() Original publisher: [Read original article]() Author: Paul Iusztin Published: 2026-06-23T08:57:02Z Content type: tutorial Language: en Sources: [Decoding ML]() Topics: [Development](), [AI Agent](), [Ground truth / benchmark quality](), [dataset]() Tags: [agent](), [ai](), [ai-evals](), [case-study](), [development](), [evaluation](), [experiment](), [saas](), [test](), [tests]() ### AI overview This case study explains Evaluation-Driven Development (EDD) for AI agents: measure a new feature, compare results before and after changes, and detect regressions before merging. It also discusses generating realistic test data when historical datasets, traces, or ground truth are unavailable. ### Source excerpt Turn every AI agent change into a measured experiment you compare before and after to detect regressions and measure performance. ## Developing a Zephyr IoT app with AI DevFeed: [Developing a Zephyr IoT app with AI]() Original publisher: [Read original article]() Author: John Lee Published: 2026-06-15T00:00:00Z Content type: tutorial Language: en Sources: [Blog on Developer Portal]() Topics: [AI Agent](), [Embedded Software Dev](), [Internet of things](), [Artificial Intelligence](), [cursor](), [Testing](), [C](), [Git](), [Open Source](), [real-time]() Tags: [agent](), [ai](), [ai-agent](), [blog](), [code](), [coding](), [cursor](), [development](), [drivers](), [embedded](), [esp](), [esp-idf](), [espressif](), [experiment](), [firmware](), [git](), [how-to](), [iot](), [llm](), [m5stack](), [testing](), [workflow](), [zephyr]() ### AI overview This article describes applying an AI-agent workflow to develop a Zephyr-based IoT application on the ESP DualKey kit from M5Stack. It emphasizes specifications, project boundaries, Git discipline, testing, and bench validation while moving the implementation from ESP-IDF and Rust to Zephyr C. ### Source excerpt Still on improving our AI-based, embedded software workflow, this article applies the same AI-agent discipline to Zephyr. Using the same ESP Dualkey kit from M5Stack, specs, integration rules, plan->execute->commit->test, reusable modules, journals, and Git - without redefining the product. ## Kotlin Multiplatform in Production: Two Real-World Use Cases from Booking.com DevFeed: [Kotlin Multiplatform in Production: Two Real-World Use Cases from Booking.com]() Original publisher: [Read original article]() Author: Diego Gómez Olvera Published: 2026-06-05T15:09:18Z Content type: article Language: en Sources: [Booking.com Development - Medium]() Topics: [Kotlin Multiplatform](), [compose-multiplatform](), [experiments](), [A/B Testing](), [Android](), [iOS](), [Design system](), [Mobile](), [Development]() Tags: [a-b-testing](), [android](), [booking](), [bookingcom](), [compose](), [compose-multiplatform](), [concepts](), [consistency](), [data](), [development](), [engineering](), [experiment](), [experimentation](), [experiments](), [ios](), [java](), [jetpack-compose](), [kotlin](), [kotlin-multiplatform](), [mobile](), [multiplatform](), [objective-c]() ### AI overview This article describes two Booking.com engineering use cases for Kotlin Multiplatform and Compose Multiplatform: a shared experimentation library for consistent experiment assignments across Android and iOS, and hosting an Android design system in a web browser. ### Source excerpt Introduction For the majority of Booking.com travelers, mobile is the primary channel for researching, planning, and booking trips. Recent data shows that over 80% of travelers rely on a mobile app during the research phase, with more than half of all bookings occurring on mobile devices. Consequently, the Android and iOS platforms are critical to the company's product strategy; engineering choices made here have significant repercussions for the entire organisation. To maintain agility at this scale, two elements must function in unison: Strict decision validation: At any time, Booking.com manages over 1,000 simultaneous experiments across its product suite, with hundreds active on mobile. Every minor adjustment undergoes A/B testing via our proprietary experimentation library before reaching the user. A unified design system ensures product consistency and makes design goals transparent to all contributors, not just maintenance engineers. This article examines two specific engineering challenges solved using Kotlin Multiplatform (KMP) and Compose Multiplatform (CMP): Developing a shared experimentation library to ensure uniform experiment assignments across Android and iOS. Using Compose Multiplatform to host our Android design system in a web browser, bridging the gap between design concepts and implementation. While both cases use the same underlying technology, each provides unique insights into multiplatform development. Use case 1: shared experimentation library on Android and iOSThe problem with two implementations Historically, our internal experimentation library, responsible for managing experiment assignments, evaluations, and tracking on mobile, was maintained as two distinct codebases: a mix of Java and Kotlin for Android and Objective-C for iOS. While intended to be identical, managing two languages with fluctuating team resources inevitably led to logic drift. Discrepancies in event-tracking and experiment-fetching behaviours emerged, though they wer ## Re-autoresearching MSMARCO BM25, on Vespa DevFeed: [Re-autoresearching MSMARCO BM25, on Vespa]() Original publisher: [Read original article]() Author: andreer thomas Published: 2026-05-29T00:00:00Z Content type: article Language: en Sources: [Vespa Blog]() Topics: [Benchmark](), [dataset](), [Python](), [Large Language Model](), [generalization in machine learning](), [pandas](), [Google Search]() Tags: [benchmark](), [bm25](), [dataset](), [evaluation](), [experiment](), [google-search](), [information-retrieval](), [openai](), [pandas](), [python](), [ranking](), [retrieval](), [zero-shot]() ### AI overview This article reproduces an MSMARCO BM25 autoresearch experiment in Vespa. It compares LLM-driven Python reranking with an approach restricted to existing Vespa rank features and reports a comparable improvement on a 650,000-passage subset, with better generalization to the full dataset. ### Source excerpt BM25 is having a moment. We reproduce Doug Turnbull's MSMARCO autoresearch experiment in Vespa and get a comparable MRR@10 lift from existing rank features -- with twice the generalization to full MSMARCO. ## Four ways Google Research scientists have been using Empirical Research Assistance DevFeed: [Four ways Google Research scientists have been using Empirical Research Assistance]() Original publisher: [Read original article]() Published: 2026-04-29T21:07:00Z Content type: article Language: en Sources: [The latest research from Google]() Topics: [Google](), [Artificial Intelligence](), [Data Mining & Modeling](), [datasets](), [Benchmark](), [Monitoring](), [real-time]() Tags: [accelerate](), [ai](), [benchmark](), [data](), [data-mining-modeling](), [experiment](), [general-science](), [generate](), [generative-ai](), [go](), [google](), [insights](), [machine-intelligence](), [monitoring](), [real-time](), [research](), [science](), [software]() ### AI overview Google Research scientists are using Empirical Research Assistance (ERA) to develop expert-level empirical software and explore AI-assisted scientific discovery across epidemiology, geospatial analysis, cosmology, atmospheric monitoring, and neuroscience. The article highlights ERA's use in computational modeling, interpretable solutions, and real-time forecasts for COVID-19, influenza, and RSV. ### Source excerpt Data Mining & Modeling [Next page]()