--- title: "Latest reads" description: "The engineering internet, summarised so you can actually read it." --- # Latest reads > The engineering internet, summarised so you can actually read it. ## Articles ### [Public AI on Hugging Face Inference Providers 🔥](https://yomu.fyi/post/public-ai-on-hugging-face-inference-providers.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Joseph Low, Joshua Tan, Célina Hanouti, Julien Chaumond, Simon Brandeis, Lucain Pouget - Published: Sep 17, 2025 - Type: Announcement Hugging Face has integrated Public AI as a supported Inference Provider on the Hugging Face Hub. Public AI operates as a nonprofit, open-source project providing access to sovereign and public models from institutions such as the Swiss AI Initiative and AI Singapore. Its distributed infrastructure combines a vLLM-powered backend serving OpenAI-compatible APIs across partner-donated clusters with a global load-balancing routing layer. Users can access these models through the Hugging Face web UI, Python client SDK, and JavaScript SDK using either direct provider API keys or routed Hugging Face tokens. At the time of announcement, inference through the Public AI provider is free of charge, supported by donated GPU time and advertising subsidies. ### [Taming the monorepo beast: Our journey to a leaner, faster GitLab repo](https://yomu.fyi/post/taming-the-monorepo-beast-our-journey-to-a-leaner-faster-gitlab-repo.md) - Company: [Grab](https://yomu.fyi/company/grab.md) - Author: Nagendra Gangwar - Published: Sep 16, 2025 - Type: Problem & solution Grab’s decade-old Go monorepo grew to 12.7 million commits, 73GB of Git-tree metadata, 176GB of blobs, 12 million references, and 444,000 files, causing replication lag, slow developer workflows, and GitLab operational bottlenecks. A staging proof of concept compared full and single-commit histories under production-like load: replication delay fell from 160–240 seconds to 1–2.5 seconds, nearly a 100x improvement. Because custom migration tooling processed about 100 commits per hour, the team retained commits reachable from more than 2,000 critical tags, one month of recent history, and simplified merge commits, while archiving older history. The resulting prune reduced commits by 99.9%, improved replication performance by 99%, preserved code and tags without data loss, and left teams with minor migration adjustments. ### [How we built it: Real-time analytics for Stripe Billing](https://yomu.fyi/post/how-we-built-it-real-time-analytics-for-stripe-billing.md) - Company: [Stripe](https://yomu.fyi/company/stripe.md) - Author: Reed Trevelyan - Published: Sep 16, 2025 - Type: Problem & solution Stripe built a real-time streaming analytics system for Billing after customers said they needed to detect emerging behavior and adapt pricing quickly, replacing a batch pipeline whose subscription updates lagged by 24 hours on average. The architecture uses Apache Flink to maintain compressed subscription history and incrementally process analytics events, while a custom Apache Spark job generates initial Flink state from billions of historical events and supports validation and exports. A new Apache Pinot v2 query engine performs windowed aggregations and joins without offline preaggregation; most updates are processed in under one minute, nearly all reach users within 15 minutes, and production query latency is under 300 milliseconds. Metric-definition changes trigger historical recalculation while new events are buffered and later replayed, preserving a consistent Dashboard view during the transition. ### [\`LeRobotDataset:v3.0\`: Bringing large-scale datasets to \`lerobot\`](https://yomu.fyi/post/lerobotdataset-v3-0-bringing-large-scale-datasets-to-lerobot.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Francesco Capuano, Michel Aractingi, Quentin Lhoest, Caroline Pascal, Pepijn Kooijmans, Jade Choghari, Remi Cadene, Simon Alibert, Adil Zouitine, Martino Russi, Steven Palma - Published: Sep 16, 2025 - Type: Announcement LeRobotDataset:v3.0 is a new release for lerobot, Hugging Face’s robotics library, addressing file-system limitations encountered when LeRobotDataset:v2 stored one episode per file at million-episode scale. Version 3 packs multiple episodes into shared Parquet and MP4 files, while relational JSON and chunked episode metadata identify episode boundaries and locate the corresponding tabular and visual data. The format separates tabular data, visual data, and metadata; it stores low-dimensional readings in Apache Parquet, concatenates camera frames into MP4 videos, and supports native temporal windowing through delta\_timestamps. It also introduces StreamingLeRobotDataset for processing batches from the Hugging Face Hub without downloading the full dataset locally, and provides a one-line conversion utility for datasets in the LeRobotDataset format. The release is planned for lerobot-v0.4.0 and is presented as a step toward datasets with millions of episodes. ### [20 Social Media Holidays to Celebrate This October](https://yomu.fyi/post/20-social-media-holidays-to-celebrate-this-october.md) - Company: [Brandwatch](https://yomu.fyi/company/brandwatch.md) - Author: Ksenia Newton - Published: Sep 15, 2025 - Type: Explainer October offers marketers a dense set of social media holidays, beginning with Global Diversity Awareness Month and continuing through Halloween. The calendar pairs each observance with a date, hashtag suggestions, and a possible engagement idea, ranging from cultural traditions and coffee orders to teacher memories, taco votes, and costume sharing. It also includes awareness-focused occasions such as International Day of Non-Violence, World Sight Day, World Mental Health Day, Global Handwashing Day, World Food Day, and United Nations Day. Several entries supply factual context, including the 2.2 billion people with visual impairment, World Vegetarian Day’s 1977 establishment, and Global Handwashing Day’s first celebration in 2008. The closing guidance recommends using these observances to craft and optimize an October content calendar. ### [Toyota, Affirm, Mercury, and more get faster business results with certified Notion Solutions Partners](https://yomu.fyi/post/toyota-affirm-mercury-and-more-get-faster-business-results-with-certif.md) - Company: [Notion](https://yomu.fyi/company/notion.md) - Published: Sep 15, 2025 - Type: Explainer Notion’s certified Solutions Partners help organizations replace fragmented tools, connect workflows, and build systems that can scale. At Affirm, Notion State used a five-phase deployment strategy across multiple departments, helping achieve 95% knowledge-management adoption, reduce tool seats, and create a unified workspace. NorthSand helped Toyota’s Future Creation Center add custom automations, approval workflows, and centralized scheduling, making key workflows three times faster and cutting meeting-preparation time in half. MegazoneCloud supported JOYCITY’s migration from Jira, Confluence, and Microsoft by running parallel systems, focusing new projects in Notion, and providing on-site training, while Optemization archived more than 63,000 of Mercury’s pages and established a standardized information architecture. The examples position partner expertise and ongoing support as ways to improve adoption, trust, collaboration, and scalability. ### [Usage Policy update](https://yomu.fyi/post/usage-policy-update.md) - Company: [Anthropic](https://yomu.fyi/company/anthropic.md) - Published: Sep 15, 2025 - Type: Announcement Anthropic is updating its Usage Policy to reflect product changes, user feedback, regulatory developments, evolving AI risks, and enforcement priorities, with changes taking effect September 15, 2025. The update adds explicit prohibitions on malicious computer, network, and infrastructure compromise while continuing to permit vulnerability discovery with system-owner consent, and provides supplementary guidance for agentic use. It narrows political restrictions from broad bans on lobbying and campaign content to prohibitions on deceptive or disruptive democratic activity and voter or campaign targeting. Law-enforcement wording is clarified without changing the permitted scope, while high-risk requirements are specified for consumer-facing legal, financial, and employment uses, including human oversight and AI disclosure. Anthropic describes the policy as a living document and says it will continue evaluation with policymakers, experts, and civil society. ### [Visible Watermarking with Gradio](https://yomu.fyi/post/visible-watermarking-with-gradio.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Margaret Mitchell - Published: Sep 15, 2025 - Type: Announcement Generative artificial intelligence tools produce images, video, audio, and text that have become increasingly difficult to distinguish from real-world media captures. To improve synthetic media transparency, Hugging Face introduced visible watermarking capabilities directly into the Gradio web application framework. Developers can now overlay visual watermarks on image and video outputs by specifying a single watermark parameter using file paths, open image objects, or NumPy arrays. The implementation also supports QR code watermarks, which can match the visual style of generated media while conveying detailed background information. Furthermore, the Gradio Chatbot component accepts a text watermark parameter that automatically attaches attribution metadata whenever users copy generated responses to their clipboards. ### [Improving Cursor Tab with online RL](https://yomu.fyi/post/improving-cursor-tab-with-online-rl.md) - Company: [Cursor](https://yomu.fyi/company/cursor.md) - Author: Jacob Jackson, Phillip Kravtsov & Shomil Jain - Published: Sep 12, 2025 - Type: Explainer Cursor Tab predicts developers’ next actions across their codebases and handles more than 400 million requests per day, creating a large stream of accepted and rejected suggestions for training. The post describes replacing a separate acceptance filter with policy-gradient reinforcement learning, allowing the model to learn when to suggest and when to remain silent while reusing its learned code representation. An illustrative reward assigns 0.75 to an accepted suggestion, -0.25 to a rejected one, and 0 to showing nothing, targeting suggestions whose estimated acceptance chance is at least 25%; the production reward also considers suggestion size and jumps to other code locations. Because policy-gradient updates require on-policy data, each new checkpoint is rolled out to users; the current deployment-and-collection cycle takes 1.5 to 2 hours, and the resulting model makes 21% fewer suggestions with a 28% higher accept rate. ### [Strengthening safeguards with US CAISI and UK AISI](https://yomu.fyi/post/strengthening-safeguards-with-us-caisi-and-uk-aisi.md) - Company: [Anthropic](https://yomu.fyi/company/anthropic.md) - Published: Sep 12, 2025 - Type: Explainer Anthropic describes an ongoing partnership with the US Center for AI Standards and Innovation (CAISI) and UK AI Security Institute (AISI) to test and improve AI safeguards during model development. The organizations evaluated multiple iterations of Constitutional Classifiers on models including Claude Opus 4 and 4.1, using pre-deployment prototypes, varied system configurations, documentation, and classifier scores for red-teaming. Their testing exposed prompt-injection bypasses, universal jailbreaks, cipher-based and input/output obfuscation attacks, and automated attack refinement that produced an effective universal jailbreak. Anthropic patched specific vulnerabilities and restructured safeguard architecture, improved detection and filtering, and used findings to strengthen evaluation, deployment monitoring, and response practices. The collaboration also showed that sustained access, real-time data, daily technical communication, and complementary government, expert, and bug-bounty testing can reveal both common exploits and subtle attack vectors. ### [Introducing the Palmyra-mini family: Powerful, lightweight, and ready to reason!](https://yomu.fyi/post/introducing-the-palmyra-mini-family-powerful-lightweight-and-ready-to.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Rakshith, Tom Peres - Published: Sep 11, 2025 - Type: Announcement WRITER released three open models in the Palmyra-mini family ranging between 1.5B and 1.7B parameters, designed for efficient inference and specialized reasoning tasks. Built on the Qwen architecture, the release includes the standard palmyra-mini base model alongside two Chain of Thought variants, palmyra-mini-thinking-a and palmyra-mini-thinking-b. In benchmark evaluations, palmyra-mini achieved 52.6% on Big Bench Hard, while palmyra-mini-thinking-a reached 82.87% on GSM8K and palmyra-mini-thinking-b reached 92.5% on AMC23. For palmyra-mini-thinking-b, applying reinforcement learning fine-tuning to an OpenReasoning-Nemotron-1.5B base improved single-shot pass@1 accuracy while reducing sampling diversity and majority@64 performance. The models are available in GGUF and MLX-BF16 quantizations and support inference engines including vLLM, SGLang, TRTLLM, and TGI. ### [Seamlessly connect Stripe events to your frontend](https://yomu.fyi/post/seamlessly-connect-stripe-events-to-your-frontend.md) - Company: [Stripe](https://yomu.fyi/company/stripe.md) - Author: Josh Kahn - Published: Sep 11, 2025 - Type: Explainer Stripe events can connect backend state changes to responsive frontend interfaces for payments, subscriptions, connected accounts, and Stripe Terminal workflows. The post contrasts polling, which can increase traffic, latency, and delta-computation complexity, with push subscriptions, while noting that a hybrid of initial API loading and incremental updates is common. It describes webhook endpoints and Amazon EventBridge as ingestion choices, then emphasizes idempotency, ordering, duplicate detection, and failure handling. Frontend delivery options include WebSockets, MQTT, gRPC, Server-Sent Events, polling, AWS AppSync, and Momento, selected according to messaging pattern, direction, latency, scale, and operational trade-offs. An abbreviated lounge example uses a PaymentIntent and Terminal reader, routes payment\_intent.requires\_payment\_method through EventBridge to AWS AppSync Events, and lets a React client mark the reader ready to collect payment. ### [Workflows: Automate compliance safeguards by flagging transactions](https://yomu.fyi/post/workflows-automate-compliance-safeguards-by-flagging-transactions.md) - Company: [Stripe](https://yomu.fyi/company/stripe.md) - Author: Ashley Ansari - Published: Sep 11, 2025 - Type: Tutorial Stripe Workflows can automate additional compliance safeguards by flagging transactions that need review before order processing. The described pattern starts when a Payment intent succeeded event occurs, retrieves the customer, and checks whether the customer’s custom\_status metadata is not “Approved” while the payment amount exceeds $5,000. When these conditions are met, the workflow emails one or more designated team members with instructions for assessing the order. The examples motivating review include unusually high order values, unfamiliar sales regions, and behavioral patterns commonly associated with fraud, such as a new customer order above a monetary threshold. Workflows uses a visual building interface in the Stripe dashboard to sequence actions across multiple Stripe products, helping enforce business rules, reduce manual effort, and support risk management. ### [Tricks from OpenAI gpt-oss YOU 🫵 can use with transformers](https://yomu.fyi/post/tricks-from-openai-gpt-oss-you-can-use-with-transformers.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Aritra Roy Gosthipaty, Sergio Paniego, Vaibhav Srivastav, Pedro Cuenca, Arthur Zucker, Nathan Habib, Cyril Vallez - Published: Sep 11, 2025 - Type: Announcement To support OpenAI's GPT-OSS series of models, the transformers library introduced several performance upgrades that apply across supported architectures. The release integrates zero-build custom kernels downloadable directly from the Hub, reducing external dependency bloat and compilation friction for operations like Liger RMSNorm and MegaBlocks MoE. Native support for MXFP4 quantization groups vector elements into 32-value blocks with shared scales, allowing GPT-OSS 20B to fit in roughly 16 GB of VRAM and GPT-OSS 120B in roughly 80 GB. In addition, transformers incorporates Flash Attention 3 with attention sinks, continuous batching via the generate\_batch API for experimentation, and automatic memory pre-allocation to speed up model loading on GPUs. ### [Fine-tune Any LLM from the Hugging Face Hub with Together AI](https://yomu.fyi/post/fine-tune-any-llm-from-the-hugging-face-hub-with-together-ai.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Zain Hasan, Artem Chumachenko, Egor Timofeev, Max Ryabinin - Published: Sep 10, 2025 - Type: Announcement Together AI and Hugging Face have introduced an integration enabling developers to fine-tune compatible Hugging Face Hub models directly on Together AI's managed infrastructure. To launch a fine-tuning job via the Python SDK, users provide a base model from Together's catalog as a configuration template alongside the target Hugging Face repository identifier. This base model template dictates GPU allocation, memory configuration, training pipelines, and inference setup for custom models with matching architectures and sizes. The workflow operates bidirectionally, pulling from public or token-authenticated private repositories and optionally pushing completed checkpoints back to the Hub upon completion. Geared toward CausalLM models under 100 billion parameters, the capability enables faster iteration cycles and domain adaptation without custom DevOps infrastructure. ### [Securing Salesforce Integrations with Heroku AppLink](https://yomu.fyi/post/securing-salesforce-integrations-with-heroku-applink.md) - Company: [Heroku](https://yomu.fyi/company/heroku.md) - Author: David Baliles - Published: Sep 10, 2025 - Type: Explainer Salesforce integrations commonly use Connected Apps or External Services, but long-lived OAuth tokens and public vendor endpoints can expose org data to replay and misuse. Heroku AppLink provides a managed bridge in which a service mesh admits requests only from explicitly connected Salesforce orgs, validates context, and exposes services as native Salesforce actions. For calls back to Salesforce, its SDK obtains short-lived, scoped credentials at runtime through user, user-plus, or authorized-user modes, avoiding refresh-token storage in vendor systems. The mesh uses the x-client-context header to convey org and user information, while the SDK keeps bearer credentials out of application code. The post describes setup commands for the AppLink add-on, service-mesh buildpack, External Service publication, and authorizations, concluding that AppLink improves governance, least privilege, and replay resistance. ### [Jupyter Agents: training LLMs to reason with notebooks](https://yomu.fyi/post/jupyter-agents-training-llms-to-reason-with-notebooks.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Baptiste Colle, Hanna Yukhymenko, Leandro von Werra - Published: Sep 10, 2025 - Type: Problem & solution Small language models often struggle to compete with large frontier models on complex, agentic data science tasks. To improve notebook-based reasoning, researchers simplified agent scaffolding down to roughly two hundred lines of code with dedicated execution and final answer tools, boosting baseline easy accuracy on the DABStep benchmark from 44.4 percent to 59.7 percent. They constructed a curated training dataset by deduplicating two terabytes of Kaggle notebooks, automatically fetching five terabytes of linked datasets, and scoring educational value and relevance using Qwen3-32B. Question-answer pairs grounded in verified execution traces were generated to fine-tune compact Qwen3-4B thinking and instruct models. The team released the trained models alongside the dataset and execution sandboxes, establishing a foundation for reinforcement learning and distillation on notebook workflows. ### [The Region Consolidation Project](https://yomu.fyi/post/the-region-consolidation-project.md) - Company: [Fly.io](https://yomu.fyi/company/fly-io.md) - Published: Sep 9, 2025 - Type: Announcement Fly.io is consolidating its region map and deprecating 17 regions across North America, Europe, Asia-Pacific, South America, and Africa. The announcement lists 17 deprecated locations, including Madrid, Bucharest, Warsaw, Atlanta, Boston, Denver, Guadalajara, Miami, Phoenix, Querétaro, Seattle, Montreal, Bogotá, Ezeiza, Rio de Janeiro, Santiago, and Hong Kong, and names 18 consolidated regions: seven in North America, five in Europe, four in Asia-Pacific, and one each in South America and Africa. For apps migrated by Fly.io, most moves should be seamless, although applications with region-specific logic must be updated to use the new region. Manual moves generally require changing primary\_region in fly.toml, scaling up in the destination with fly scale count, and scaling down in the original region. If volumes are involved, the guidance recommends stopping existing machines, forking volumes into the new region, and creating machines from those forks; support is offered for issues that arise. ### [mmBERT: ModernBERT goes Multilingual](https://yomu.fyi/post/mmbert-modernbert-goes-multilingual.md) - Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md) - Author: Marc Marone, Orion Weller, William Fleshman, Eugene Yang, Dawn Lawrie, Ben Van Durme - Published: Sep 9, 2025 - Type: Announcement mmBERT is a massively multilingual encoder model trained on more than 3T tokens across over 1,800 languages to improve upon existing multilingual architectures like XLM-R. Built upon ModernBERT with a Gemma 2 tokenizer, mmBERT employs a three-phase training curriculum consisting of pre-training on 60 languages, mid-training on 110 languages, and a final decay phase covering 1,833 languages. The training pipeline integrates an inverse mask ratio schedule, dynamic language temperature annealing, and TIES merging across three decay variants. Benchmark evaluations demonstrate strong natural language understanding on English GLUE and multilingual XTREME, as well as competitive retrieval performance on MTEB v2 and CoIR. The release includes standard base and small models alongside open training data and checkpoints. ### [Triage and Fix with Confidence: heroku run and OTel on Heroku Fir](https://yomu.fyi/post/triage-and-fix-with-confidence-heroku-run-and-otel-on-heroku-fir.md) - Company: [Heroku](https://yomu.fyi/company/heroku.md) - Author: Su Glasgo - Published: Sep 8, 2025 - Type: Explainer Heroku Fir addresses production troubleshooting when a release causes failures, such as a background worker breaking after a database schema change, without requiring access to a live formation dyno. Its interactive heroku run command launches a dedicated one-off dyno with the application code and environment variables, relaying terminal input and output for tasks such as rake db:migrate; using it requires a public SSH key and a deployed application. The platform’s native OpenTelemetry (OTel) signals add router request-duration summaries and percentiles, represent heroku.router.connect and heroku.router.service as floats, and attach heroku.app.name to application signals. Together with service.name filtering for first-party services, these capabilities support identifying affected applications, applying isolated fixes, and verifying latency and errors through an observability platform. [Newer posts](https://yomu.fyi/posts/page/171.md) · [Older posts](https://yomu.fyi/posts/page/173.md)