# blog-posts Published articles for blog-posts. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## From Traditional ML to AI Agents: How Booking.com Scales AI Observability With Arize AI DevFeed: [From Traditional ML to AI Agents: How Booking.com Scales AI Observability With Arize AI]() Original publisher: [Read original article]() Author: Amir Bitaraf Published: 2026-07-10T07:52:18Z Content type: article Language: en Sources: [Booking.com Data Science]() Topics: [ai observability](), [agent observability](), [Machine learning](), [observability](), [human review]() Tags: [agents](), [ai](), [ai-agents](), [ai-observability](), [blog-posts](), [data-quality](), [human-review](), [latency](), [ml](), [observability]() ### AI overview Booking.com describes building an AI-native observability stack for traditional machine learning systems and agentic AI workflows. The article explains that its diverse systems require observability to detect changes, regressions, data quality issues, misconfigurations, and responsible-AI concerns across different operating constraints and user contexts. ### Source excerpt Building an AI-native observability stack for agentic AI and traditional ML at Booking.com Authors: Amir Bitaraf, Shahaf Veber Why AI Observability Matters at Booking.com At Booking.com, AI helps travellers and partners in every step of their journey, from how people discover destinations to the way we support them while they're on the road. Rather than a single flagship model, we rely on a large and growing collection of systems that each solve a specific problem at scale. To make this concrete, consider a few examples: Trip planning assistants that help travelers turn vague ideas ("somewhere warm in April with good hiking") into concrete, bookable itineraries. On-site helpers that turn property details, amenities, reviews, and options into plain-language guidance, so people can choose the right stay with confidence. Partner copilots that help accommodation partners and other suppliers respond to guest messages faster and more consistently, while still staying in control of the final reply. Ranking systems that decide which options to show first in search and recommendation to surfaces, balancing user relevance with experimentation needs. Fraud detection models that quietly protect customers and partners in the background by flagging suspicious activity before it turns into real harm. Each of these systems is built and iterated on by different teams, uses different data, and runs under different constraints such as real-time vs batch, strict latency budgets vs more relaxed ones, fully automated vs human-in-the-loop. As we scale this ecosystem, observability becomes a first-class requirement, not a nice-to-have as we need to: Know when something changes in the real world, a new travel pattern, a data quality issue, a misconfiguration and how that affects model behaviour and user experience. Detect regressions early: slower responses, more confusing answers, drops in relevance or conversion, or subtle shifts that only show up for specific geographies, devices, or use ## JulyReply - Re: Stuck in draft DevFeed: [JulyReply - Re: Stuck in draft]() Original publisher: [Read original article]() Author: Harsh Shandilya Published: 2026-07-09T18:52:00Z Content type: opinion Language: en Sources: [Posts on Harsh Shandilya]() Topics: [Learning](), [Parser](), [REST API](), [Containers](), [Git](), [Cloudflare](), [Compose](), [LLMs](), [Command-line interface]() Tags: [blog](), [blog-posts](), [blogging](), [cli](), [cloudflare](), [compose](), [containers](), [git](), [julyreply](), [llms](), [rest-api](), [technical-writing]() ### AI overview The author reflects on managing unfinished blog drafts, explaining that feedback, public interest, curiosity, and motivation sometimes lead to completion while many drafts are abandoned. They invite readers to vote on several proposed technical topics, including Retrofit, an HTML parser, Git, Docker containers in NixOS, Cloudflare, Compose UI, and LLM-assisted development. ### Source excerpt On drafts and being okay with them ## Booking.com 2026 GenAI and ML PhD Research Internship in Amsterdam DevFeed: [Booking.com 2026 GenAI and ML PhD Research Internship in Amsterdam]() Original publisher: [Read original article]() Author: Yang Yang Published: 2026-02-05T10:39:26Z Content type: article Language: en Sources: [Booking.com Data Science]() Topics: [Machine Learning & Artificial Intelligence](), [genai](), [AI Research](), [Python](), [Large Language Model](), [Apache Spark](), [SQL](), [Hadoop](), [Reinforcement learning](), [Synthetic Data Generation](), [Transformer]() Tags: [2026](), [ai](), [big-data](), [blog-posts](), [featured](), [genai](), [internship](), [machine-learning](), [ml](), [python](), [research]() ### AI overview Booking.com is recruiting current PhD students in quantitative fields for a three-month GenAI and machine learning research internship in Amsterdam in 2026. Projects include LLM alignment, transformer explainability, embeddings, context engineering, and synthetic data generation. ### Source excerpt At Booking.com, we don't just use Machine Learning -- we use it to solve some of the most complex travel challenges in the world. We're looking for the next generation of researchers to join our Machine Learning community in Amsterdam for a 3-month deep dive into cutting-edge AI. The Program As a Research Intern, you'll be embedded in our teams, working alongside world-class mentors. Your mission? To tackle real-world problems and push the boundaries of the state-of-the-art. Are You the One? We're looking for current PhD students in quantitative fields (CS, Math, AI, Physics) who can conduct independent research and have a solid grip on Python and Big Data tech (SQL, Spark, Hadoop). What's in it for you? You won't just be "an intern". You'll be a contributor to our Machine Learning community. You'll have the opportunity to contribute to the existing efforts of the Machine Learning teams, participate in internal knowledge-sharing sessions, and enjoy the collaborative, high-energy environment of our Amsterdam HQ. Projects Regularized Target Encoding for large real-world datasets Multi-Agent Collaboration Aligning LLMs with user feedback via reinforcement learning Multi-level treatments Interpretable Foundations: Explainability Methods for Transformer Models on Sequential Event Data Scalable and generalisable ID embedding learning Improving property embeddings with better handling of rich and long-context data Utility-aware retrieval for context engineering in travel planning Synthetic Data Generation in Images Requirements We are looking for independent researchers with strong understanding of Machine Learning topics (see requirements for each project in the Linkedin ad), have a track record of peer-reviewed publications and a passion for solving complex problems. Why Booking.com? You'll join a vibrant, diverse community of data scientists and researchers who love to experiment. Beyond the code, you'll experience the unique culture of our Amsterdam headquarters -- a hub ## Beyond Prompt Engineering: How We Used Supervised Fine-Tuning for Travel Recommendations DevFeed: [Beyond Prompt Engineering: How We Used Supervised Fine-Tuning for Travel Recommendations]() Original publisher: [Read original article]() Author: Amit Meitin Published: 2026-01-29T10:53:21Z Content type: article Language: en Sources: [Booking.com Data Science]() Topics: [Fine-tuning](), [recommendation systems](), [Large Language Model](), [real-time](), [AI Development]() Tags: [ai](), [blog-posts](), [engineering](), [fine-tuning](), [inference](), [language-models](), [machine-learning](), [models](), [prompt](), [recommendations]() ### AI overview Booking.com describes how it used supervised fine-tuning to improve travel recommendations for its AI Trip Planner. The approach combines large language models' understanding of unstructured traveler requests with structured behavioral data from searches, clicks, and bookings, while the article reports superior recommendation metrics and 3x faster inference. ### Source excerpt How fine-tuning delivered superior recommendation metrics while achieving 3x faster inference Every traveler is unique, and so is every trip. At Booking.com, we're always looking for new ways to make trip planning feel less like a chore and more like an adventure. That's why we set out to build an AI Trip Planner that doesn't just answer questions, but actually understands what travelers want, expressed in ways they are most comfortable with. But here's the challenge: travelers today are expressing their needs in ways that are more unstructured and nuanced than ever before. Instead of ticking boxes or picking from drop-down menus, people describe their dream vacations in their own words. Sometimes these requests are vague, sometimes very specific, but always personal. Traditional machine learning models, which thrive on structured data and clear signals, can struggle to keep up with this new level of expressiveness. At the same time, we at Booking.com have years of valuable, structured data from searches, clicks, and bookings. This data captures what travelers actually do: the trips they plan and the vacation they go on. The big question for us became how to combine the best of both worlds -- How can we harness the power of Large Language Models (LLMs) to understand and respond to unstructured requests, and use the wealth of behavioral data that has driven our recommendations for years? This post is about how we tackled that question and what we learned along the way. What is the AI Trip Planner? The AI Trip Planner is Booking.com's conversational assistant that helps travelers find destinations, accommodations and other trip components through natural, real-time chat. Instead of searching with filters, users describe what they want in their own words. The planner suggests options, whether it is inspiration or specific places to stay. Integrated into the app, it makes planning and discovering seamless and is now available in multiple countries and languages. Bridging ## AI Agent Evaluation DevFeed: [AI Agent Evaluation]() Original publisher: [Read original article]() Author: Antonio Castelli Published: 2026-01-21T13:11:23Z Content type: tutorial Language: en Sources: [Booking.com Data Science]() Topics: [LLM evaluation / benchmarking](), [AI Agent](), [Artificial Intelligence](), [Tool]() Tags: [agent](), [ai](), [ai-agent](), [api](), [best-practices](), [blog-posts](), [evaluation](), [llm-agents](), [llm-evaluation](), [machine-learning](), [sql](), [tools]() ### AI overview Booking.com's article presents practical guidance for evaluating LLM agents. It explains how agents differ from single LLMs because they can use external tools and iterate through thoughts, actions, and observations, then introduces black-box and glass-box evaluation approaches. ### Source excerpt AI Agent Evaluation: practical tips at Booking.com Authors: Zeno Belligoli, Antonio Castelli, George Chouliaras This article is the 2nd part of our GenAI evaluation best practices series. You can read the first part, focusing on single LLM evaluation, here. 1. Overview LLM agents are advanced AI systems that leverage large language models to perform complex tasks autonomously. Unlike single LLMs that respond to direct prompts, agents can break down problems, use external tools, and iterate on solutions through a series of thoughts, actions, and observations. This allows them to achieve goals that require more intricate planning, reasoning, and interaction with various environments. Agents can autonomously determine if tool utilization is necessary to fulfill a user's request. For example, they might employ: a calculator for mathematical operations, a flights API to retrieve available flights for a certain destination, or execute a SQL query to fetch information about an hotel reservation from a database. Figure1: Schematic view of an agent. The user might interact with the agent within a multi-turn conversation, asking it to perform various tasks (recommend vacation destinations, ask questions about a hotel etc.). The agent should try to complete the requested tasks abiding by a set of rules (or constraints) provided by the product specification (e.g. do not recommend properties outside the inventory). In performing the tasks the agent has freedom to use a certain set of tools, and it can interact with them via API requests. Tool examples typically range from general (e.g. calculate travel distance) to use case specific functionalities (e.g. retrieving information from a database). While these enhanced capabilities are fundamental to solve tasks which would be difficult (or even impossible) to solve with text generation only, they require a more complex evaluation process compared to evaluating a single "prompt-response" LLM. The evaluation of an agent's performance ## Ben Recht on Meehl's Philosophical Psychology DevFeed: [Ben Recht on Meehl's Philosophical Psychology]() Original publisher: [Read original article]() Published: 2024-07-27T18:49:42Z Content type: opinion Language: en Sources: [Jeremy Kun]() Topics: [Data Science](), [Math and Logic]() Tags: [blog-posts](), [mathematics](), [philosophy](), [science](), [shortform](), [social-sciences]() ### AI overview A review of Ben Recht's blog series on Paul Meehl's Philosophical Psychology, which examines philosophy of science, scientific debate, and weaknesses in statistical studies in the social sciences. The series also discusses alternatives to relying on conventional hypothesis testing, including historical experimental examples. ### Source excerpt Ben Recht, a computer science professor at UC Berkeley, recently wrapped up a 3-month series of blog posts on Paul Meehl's "Philosophical Psychology." Recht has a table of contents for his blog series. It loosely tracks a set of lectures that Meehl gave in 1989 at the University of Minnesota. In it, he surveys of the philosophy of science, lays out a framework for scientific debate, and critiques scientific practice. Recht summarizes his arguments, simplifies the ideas, provides examples, and offers his own commentary, considering today's computerized world. ## Testing Images in Blog Posts DevFeed: [Testing Images in Blog Posts]() Original publisher: [Read original article]() Published: 2024-03-16T00:00:00Z Content type: tutorial Language: en Sources: [Gaurav Sarma's Blog]() Topics: [Testing](), [Image](), [test]() Tags: [blog](), [blog-posts](), [images](), [posts](), [testing]() ### AI overview A blog post demonstrates testing different types and sizes of images, including remote images loaded from URLs and local images stored in a directory. ### Source excerpt Testing Different Types of Images Remote Image Here's a remote image from a URL: . [Remote test image](https://picsum... ## Hello Deep Learning: Reading handwritten digits DevFeed: [Hello Deep Learning: Reading handwritten digits]() Original publisher: [Read original article]() Published: 2023-03-30T10:00:04Z Content type: tutorial Language: en Sources: [Bert Hubert's writings]() Topics: [Deep learning](), [Neural Network](), [Deep neural networks](), [Image]() Tags: [blog-posts](), [deep-learning](), [images](), [layer](), [matrices](), [neural-network](), [pixels](), [spatial]() ### AI overview A tutorial in the Hello Deep Learning series explains how to build a neural network that recognizes all ten handwritten digits. It describes flattening 28x28 images, applying matrix multiplications and ReLU layers, and selecting the highest-scoring output. ### Source excerpt This page is part of the Hello Deep Learning series of blog posts. You are very welcome to improve this page via GitHub! In the previous chapter we described how automatic differentiation of the result of neural networks works. In the first and second chapters we designed and trained a one-layer neural network that could distinguish images of the digit 3 and the digit 7, and the network did so very well. ## This month in Mobian: May 2022 DevFeed: [This month in Mobian: May 2022]() Original publisher: [Read original article]() Author: Mobian team Published: 2022-05-31T23:59:59Z Content type: article Language: en Sources: [Mobian's Blog]() Topics: [Development](), [GitLab](), [ci](), [Debian](), [Kernel](), [systemd](), [Wiki]() Tags: [blog-posts](), [bugs](), [ci](), [configs](), [debian](), [gitlab](), [kernel](), [kernels](), [repositories](), [systemd](), [update](), [wiki]() ### AI overview A May 2022 Mobian development update covers monthly developer meetings, planned repository and developer-wiki moves toward Debian infrastructure, kernel and audio fixes, universal images for PinePhone devices, tow-boot limitations on PineTab, and a PinePhone Keyboard driver change. ### Source excerpt This is the first in a (hopefully) series of short blog posts summarizing what has happened in Mobianland. Given that Mobian integrates much of the ecosystem not developed by us, this report could reiterate many of the things being said somewhere else. Monthly dev meeting Despite being located all over the world (really!), we manage to meet in person once a month. Well, ok, "meet in person" means we conduct a jitsi video call. It is fun seeing all these sleepy faces where half of the people would like to go to bed already, while others wished they could have remained in bed longer. The TL;DR: we will attempt to move over more repositories from gitlab to salsa.debian.org (given gitlabs changes to the CI allowance we get, and in an attempt to align ourselves closer to Debian). In a similar move, the developer's wiki (NOT the user wiki) moves over to the Debian wiki. pondering if systemd-repart and systemd-growfs can maybe replace "growroot" We really should be opening a user-contrib section or something where contributed ports and kernels can live, e.g. for the Nexus 5. Planning stage. OnePlus 6 / Pocophone F1 Audio was broken due to a missing config option on 5.17 kernels, and was fixed by a subsequent update. Thanks to everyone who reported this issue and provided the information allowing us to fix it. Universal Images We have a universal mobian image that boots on the original Pinephone and the Pinephone Pro, however, there are still some bugs to iron out and kernel configs to be adapted. But overall, this looks promising. However, this will require some tool that can adapt configuration settings at runtime, based on the device it is currently running on. Tow-boot We were one of the earliest distributions to move over to tow-boot, and that worked out nearly well. However, there is a little catch: There is no tow-boot for the PineTab yet. So all images post-March 27, will not be able to install a U-boot if you have non installed yet. Help to port tow-boot to the Pin ## My next book will be "Practical Math for Programmers" DevFeed: [My next book will be "Practical Math for Programmers"]() Original publisher: [Read original article]() Published: 2022-03-16T07:00:00Z Content type: release Language: en Sources: [Jeremy Kun]() Topics: [Software](), [coding](), [Python](), [math](), [GitHub]() Tags: [announcements](), [blog-posts](), [book](), [code](), [code-documentation](), [github](), [math](), [practical](), [python](), [repo](), [software](), [test]() ### AI overview Jeremy Kun announces a forthcoming book, Practical Math for Programmers, about mathematics used in production software. The book is planned as an accessible survey with working Python 3 demonstrations, a GitHub repository, tests, documentation, and practitioner interviews. ### Source excerpt tl;dr: I'm writing a new book, sign up for the announcements mailing list. I've written exactly zero new technical blog posts this year because I've been spending all my writing efforts on my next book, Practical Math for Programmers (PMFP, subtitle: A Tour of Mathematics in Production Software). I've written a little bit about it in my newsletter, Halfspace. There I rant, critique, brainstorm, and wax poetic about math and software. ## 2018 in Review DevFeed: [2018 in Review]() Original publisher: [Read original article]() Published: 2019-01-04T00:00:00Z Content type: opinion Language: en Sources: [Una Kravets]() Topics: [Development](), [Web Development](), [CSS](), [User interface design](), [Design system](), [Bulma](), [gatsby]() Tags: [blog-posts](), [css](), [design-systems](), [dev](), [flexbox](), [projects](), [review](), [speaking](), [ui]() ### AI overview A personal review of 2018 covering health and family experiences, work at Bustle Digital Group, podcast and blog production, side projects including Gatsby and Flexbox work, and conference speaking. ### Source excerpt Reviewing how my 2018 went, and discussing goals for next year! ## 2015: a year in review DevFeed: [2015: a year in review]() Original publisher: [Read original article]() Author: Monica Dinculescu Published: 2015-12-22T00:00:00Z Content type: opinion Language: en Sources: [Monica Dinculescu]() Topics: [Chromium](), [JavaScript](), [Internet](), [Web platform](), [C++](), [browser](), [GitHub]() Tags: [blog-posts](), [c-plus-plus](), [chromium](), [github](), [javascript](), [push-notifications](), [web-platform](), [year-in-review]() ### AI overview A personal year-in-review describing burnout after the launch of Chrome's Profiles UI, the author's departure from Chromium, and joining Polymer to work on JavaScript and the web platform. The author also reflects on projects, talks, and lessons from the year. ### Source excerpt I've never really done a year in review. One day, I'd like to open source my goals, but since I'm still a chicken, this is a baby step towards that. Plus, this is one of the first years I'm really proud of, and things that you're proud of tend to live on the Internet, for posterity. Here's what my GitHub contributions say about it: Burning out The year started off really poorly. My team had just shipped the new Profiles UI in Chrome, after a year and a half of hard work, and it was met with a looooot of Internet anger. On one side we had data to prove that the change we did was right, which made the powers that be want to stick by it; on the other side I had Twitter, who was calling me names and wanting me fired. Kind of ironic, since I was just the person who implemented the feature and had no power to change it. I think what burnt me out wasn't waking up to a stream of negative emails and tweets, it was knowing that there was absolutely nothing I could do about it other than wait. So I started working on dumb side projects to feel better. I made a link aliaser. I bought dumb domains. I wrote blog posts about the only thing that I knew, which was working on Chromium. I noticed that not working on Chromium made me happy. So I bit the bullet, left Chromium, and joined Polymer. Joining Polymer Looking back, I picked Polymer for a bunch of silly reasons that ended up working out spectacularly well. I wanted to leave Montreal. I wanted to work on JavaScript, since it was the only thing keeping me going. I didn't want to commute to Mountain View, which reduced my options by like a billion percent, and I wanted to ship things. Polymer had all of that. So on April 15, I packed my cat and my books and moved to San Francisco. Polymer is my dream job. I get to write code that I'm genuinely passionate about. I get to try to change the web platform, and talk about why I think we're doing the right thing. Most importantly, I get to ship something everyday. It turns out that's a