Loading…
Latest reads
Every post, summarised
Grounded summaries of engineering blogs, newest first. Each links to the original.
- 4,671 posts
- 20 companies
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Latest posts
Tech – Tools & craft – Notion
Notion’s engineering collection covers tools, product design, infrastructure, security, and data systems across a set of technical posts. It opens with an inside look at the hosted Notion MCP server, created in response to requests for AI agents to work directly with Notion workspaces; the path described runs from an open-source experiment to a hosted solution with custom tools for AI workflows. Other entries describe Lore’s conversion of conversations, decisions, tasks, and durable facts into Notion pages, a Spark-native lexical-search reindexer, and multi-region systems for search, analytics, AI, and event logging. The collection also reports CI guardrails for schema changes, security testing for Custom Agents, vector-search scaling by 10x with costs reduced by 90%, Spot Balancer for Spark on Kubernetes, and the architecture and data model behind Offline Mode.
Build Low-Latency Multilingual Voice Agents: Open Weights & Full Deployment Control with NVIDIA Magpie TTS
NVIDIA Magpie Multilingual TTS is a 364M-parameter open-weights model for multilingual voice applications, supporting 12 languages, including newly added Modern Standard Arabic, Korean, and Brazilian Portuguese. It is positioned for cascaded voice-agent architectures in which ASR, TTS, and LLM components are independently deployed, tuned, and scaled on controlled infrastructure. Frame stacking predicts two audio frames per decoding step, while a local transformer models dependencies between simultaneously generated codebook tokens; NVIDIA reports 32–79 ms single-stream TTFA across its listed GPUs. Compared with the previous release, French and Spanish CER improved to 1.54% and 0.60%, while SSIM reached 0.747 and 0.793; German SSIM also increased to 0.742. The Hugging Face checkpoint supports research and fine-tuning, while NVIDIA NIM provides optimized production serving and NeMo supports customization.
Introducing FILE type: a native column type for multimodal data
The beta of FILE introduces a native column type for storing unstructured data such as documents, images, audio, and video alongside structured table data. FILE applies fine-grained governance, row- and column-level controls, ABAC, and lifecycle synchronization, so deleting a row also deletes the corresponding binary in object storage. Its lightweight pointers fetch file content only when a query requires it, while SQL, Python UDFs, built-in AI functions, and materialized views support multimodal processing. The example uses sampled dashcam frames and object detection to identify stops with no hazard, then joins extracted results with trip metadata; the announcement also describes connectors, incremental pipelines, and an open effort around Parquet, Delta Lake, Apache Iceberg, and Apache Spark.
Rebuilding Notion’s lexical search reindexer
Notion’s lexical search reindexer had become a multi-week, manually monitored process that produced about 90% data consistency. The old design also relied on Snowflake metadata lookups and a separate catchup pipeline. The redesign replaces ECS/Fargate workers and live Elasticsearch writes with an Apache Spark–native pipeline that routes documents to exact index-and-shard pairs and generates native snapshot files through embedded Elasticsearch nodes. Temporary indices capture live updates while the snapshot is built, after which aliases, external document versions, Elasticsearch’s _reindex, and phantom detection handle catchup, including alive-to-deleted transitions. The new system completes full reindexing in under two days, catchup in under an hour, reaches 100% consistency, removes Snowflake and ECS from the indexing path, and reduces manual intervention to under two hours per run.
Run Local Agentic AI Workflows with Meta’s Muse Glimmer on NVIDIA
Meta has released Muse Glimmer, a 30B open-weight dense model with a 120K+ context window aimed at local agentic AI workflows. The model uses a dense architecture that activates every parameter per token, avoiding routing overhead and targeting reliable instruction following, long-context coherence, and predictable latency for multi-step agent tasks. On NVIDIA Blackwell Ultra hardware, Muse Glimmer reports over 20K tokens/sec/GPU at BF16/NVF4 precision, with the full model fitting in a single GPU's VRAM. Deployment options span NVIDIA GeForce RTX 5090, DGX Spark, DGX Station, and Jetson edge devices, supported by vLLM, SGLang, and NVIDIA NIM containers. Fine-tuning is available through NVIDIA NeMo AutoModel for SFT and LoRA, with NeMo RL for reinforcement learning, and the NemoClaw harness enables long-running assistants in a sandboxed environment.
Serving the most critical missions: Cloudflare for Government achieves FedRAMP Class D (High) Certified status
Cloudflare for Government has achieved FedRAMP Class D (High) certification status, following its FedRAMP Moderate authorization in 2022 and sponsorship by the National Institute of Standards and Technology. FedRAMP High covers systems handling the nation’s most sensitive unclassified data, including law enforcement, emergency services, financial systems, and national security information, where a control breach could be catastrophic. Cloudflare says its offering runs the same software stack and services as its commercial platform on one global network, while its Data Localization Suite ensures traffic inspection and processing for FedRAMP High occur exclusively in U.S. data centers. The systems developed for FedRAMP High are also the foundation of Cloudflare’s stated pursuit of DoD Impact Level 4 authorization, with access to its Zero Trust, application performance, and developer features as they are released.
14 Social Media Holiday to Celebrate This September
The September social media calendar features 14 holidays and events that brands can use to plan content, connect with audiences, and encourage engagement. The list spans World Letter Writing Day, International Day of Charity, Labor Day, International Literacy Day, National Video Games Day, International Equal Pay Day, and International Podcast Day, among others. Each entry pairs its date and hashtag with a suggested angle, such as handwritten letters, employee recognition, literacy infographics, gaming contests, equal-pay advocacy, or sustainable travel destinations. The calendar also includes September 11’s National Day of Service and Remembrance, Hobbit Day, the First Day of Fall, and World Tourism Day, with ideas ranging from commemoration and themed visuals to seasonal recommendations and education. It closes by directing readers to a 2026 social media holiday calendar for further content-planning ideas.
Making Knowledge Distillation Cheap Enough to Run at Scale
Knowledge distillation trains a smaller student to match a larger teacher's capabilities, but conventional online training keeps both models loaded and materializes full vocabulary distributions at every step, making long-context recovery expensive. The proposed system caches the teacher's top-100 logits once and introduces a fused, chunked KL-divergence loss that avoids storing the full vocabulary-by-sequence grid or the student's full logits. In benchmarks with an 8K-token context on one H200, offline fused chunked KL used 58.3GB peak memory versus 102.8GB for online training, while all four methods reached near-identical training loss. At 32K tokens, the fully chunked loss reduced isolated kernel memory from 85.2 GiB to 5.45 GiB; distilling GPT-OSS 20B at that context cut step time from 57.0 to 12.23 seconds and reduced the setup from four GPU nodes to one.
Meta is back with Muse Glimmer: local, agentic, multimodal, and open source
Meta’s Muse Glimmer is a 30B-parameter multimodal model released under the Apache 2.0 license for local, agentic applications including coding, document analysis, and personal assistants. Its dense architecture combines a 2B ViT-style perception encoder with a 28B text decoder, while an optional DFlash speculative-decoding drafter targets faster generation for structured content. The decoder uses alternating sliding-window and full-attention layers, grouped-query attention, and query-key normalization with extra query scaling. Day-zero support is available in Transformers, llama.cpp, vLLM, Inference Endpoints, and other libraries, with examples for text, images, videos, deployment, and agent workflows. Published benchmarks show varied results: Muse Glimmer leads on MCP Atlas and SWE-Bench Pro, while Qwen3.6 leads on OSWorld-Verified and SWE-Bench Verified.
From charging a blended rate to IC++: pass network costs to your connected accounts
Platforms using blended payment rates absorb interchange variance, which can reduce margins as card mix shifts toward premium rewards, corporate, and other high-interchange cards. Network cost passthrough, also called IC++, lets eligible Connect platforms pass interchange, scheme fees, and applicable adjustments to connected accounts at cost while keeping Stripe fees and platform markup separate. Eligibility requires IC+ pricing, platform-controlled pricing, enabled transfers, an appropriate Merchant of Record fund flow, and eligible geographies; passthrough can then be enabled in the Dashboard or through the API. The rollout also requires lower application fees, new network-cost reporting, and a pilot with clear communication before broader deployment.
How Musinsa scaled its audience engine with ClickHouse Cloud and reduced TCO by 71.4%
Musinsa’s Audience Engine powers a customer data platform that targets audiences across 16.4 million members. It stores each user-to-audience relationship as a separate row, producing roughly 1.1 billion cohort-to-user mappings across 2,500 audiences and potentially 16 million new rows when one audience is added. Its self-hosted ClickHouse deployment coupled storage and compute, forced workloads to share a cluster, and created operational burdens as data and query demands grew. Musinsa migrated to ClickHouse Cloud on AWS, using separated S3-backed storage, workload-specific compute, automatic idling, vertical autoscaling, and ClickPipes for direct real-time ingestion. The company reports an 86.5% reduction in storage costs and a total cost of ownership reduction of up to 71.4%, while shifting engineering effort toward CDP business logic.
How We Moved 1,500 Android Screenshot Tests to Roborazzi
Thumbtack migrated 1,500 Android screenshot tests from Firebase Test Lab to Roborazzi to eliminate slow and error-prone testing on remote emulators. The engineering team ported test suites from androidTest to test in batches of 20, adjusting test rules and annotations for Robolectric. Because Roborazzi runs in a single JVM process with parallel execution, the team had to debug shared font cache pollution and cross-platform shadow rendering inconsistencies between macOS and Debian. They resolved configuration repetition by mapping Robolectric annotations into module-level robolectric.properties files and avoided stale cached images by clearing Roborazzi's cache via a custom Gradle task. The migration significantly decreased CI runtimes while making local test recording and verification easier.
How and Why Netflix Built a Real-Time Distributed Graph: Part 3 — Querying the graph with gRPC…
Netflix designed a query execution serving layer for its Real-Time Distributed Graph to power sub-100ms responses across diverse graph traversal workloads spanning billions of nodes and edges. To prevent compounding network delays in multi-hop queries, the engine uses a breadth-first traversal model that batches entity lookups across entire frontier levels rather than tracing paths depth-first. The entire serving architecture relies on asynchronous composition across small thread pools of 16 to 24 threads, ensuring no thread blocks while waiting on remote storage or enrichment I/O. Selective caching via EVCache targets stable properties with volatility-matched TTLs, producing 70 to 80 percent cache hit rates and reducing backend storage calls by three to four times. A layered filtering hierarchy pushes depth and edge limits directly to the storage tier, eliminating bespoke code changes while isolating system resources against excessive fan-out.
Unveiling good and bad behaviors on the Agentic Internet
Web traffic increasingly shifts between human interaction and automated agents, rendering point-in-time security checks ineffective against sophisticated automation. Cloudflare addresses this shift through continuous behavioral analysis designed to evaluate trust across entire user sessions rather than relying on static point-in-time hurdles. Its CDN-injected client-side system, Precursor, tracks continuous behavioral indicators like cursor movements to catch mid-session automated takeovers. To counter rapid bot adaptation, Cloudflare is introducing self-adjusting machine learning models under Adaptive Intelligence and non-deterministic defenses such as the AI Labyrinth. These tools allow site owners to poison scraper training data, trap crawlers in synthetic webs, or queue legitimate agents while dynamically penalizing abusive traffic.
Introducing Radar Researcher: An AI tool for exploring Internet data in plain language
Cloudflare launched the beta of Radar Researcher, an AI tool designed to let users explore global Internet traffic data using plain-language queries. Built entirely on Cloudflare's developer platform, the system uses the Cloudflare Agents SDK within a Cloudflare Worker, maintaining conversation state and SQLite databases inside Durable Objects. The inference layer relies on Workers AI running models like Kimi K2.7 through an ordered fallback chain across three model families to prevent outages, routed via AI Gateway for logging, caching, and guardrails. Radar's frontend renders retrieved data using its existing native interactive charts, while smaller dedicated models generate conversation titles and follow-up suggestions asynchronously. Additionally, Radar introduced support for the WebMCP standard, exposing imperative and declarative tools to allow external browser agents to navigate and interact with the platform directly.
Announcing Cloudflare Ambassadors, Community Engineers, and another $1M in open-source funding
Cloudflare introduced an updated community initiative featuring two tracks: Cloudflare Ambassadors and Cloudflare Community Engineers. The Ambassador program selects community leaders for terms lasting up to two years to organize events, lead student groups, and create learning spaces with company-provided credits, resources, and marketing assets. The Community Engineer track offers grants from a new one million dollar open-source fund spanning the next two years to support maintainers of projects such as Astro, Agents SDK, EmDash, Hono, and Vinext. To maintain its Discord server of nearly 100,000 members, a new Discord committee composed of staff and Ambassadors will guide community discussions while automated spam protections handle routine moderation.
Unifying Workers AI and AI Gateway into a single AI control plane
Workers AI and AI Gateway are converging into a unified AI control plane to simplify model access, billing, and observability. Developers can now access both services through a single entrypoint using unified Workers bindings and the /ai/ REST endpoint. Specifying a default gateway automatically instruments requests with payload logging, token tracking, and cost attribution without manual dashboard setup. Cloudflare has also rolled out unified prepaid billing, allowing teams to spend AI Gateway credits on Workers AI models while unlocking higher rate limits. Planned future features include model-first routing for automated provider failover and smart routing powered by task classifiers on Workers AI.
Managing AI Coding Costs at Scale
AI coding tools have improved velocity metrics at Databricks, but broad deployment creates exponentially growing costs that can threaten the efficiency gains they deliver. The post presents a cost-management playbook aimed at preserving broad, low-friction access while keeping aggregate spending within a roughly fixed per-user envelope. It emphasizes the coding-model efficiency frontier, automated evaluations of internal workloads, model-flexible harnesses or meta-harnesses, and request- or task-level routing to the cheapest capable model. Databricks reports that Unity AI Gateway Smart Router reduced average task cost by more than 30% while roughly matching the most expensive model’s quality, and that harness and caching changes cut generated tokens and associated costs by almost 50% without observed quality degradation. The proposed AI Gateway centralizes model access, budget policies, tool configuration, and coding-session traces for ongoing optimization.
What is an AI Assistant?
AI assistants are software systems that use language models, data retrieval, and reasoning to understand natural-language requests and return information or perform tasks. For enterprise data teams, they can generate SQL, build dashboards, troubleshoot pipelines, and automate repetitive work, but their usefulness depends on integration with data catalogs and governance rather than generic text generation alone. The described pipeline captures input, parses intent and context, retrieves knowledge, generates an output, and either delivers it for review or executes an action; advanced systems incorporate tool and API use plus feedback from corrections and approvals. The article categorizes assistants as voice, conversational, general-purpose, domain-specific, data and developer, or autonomous agents, while noting risks including hallucinations, privacy exposure, integration complexity, over-reliance, and bias. It presents assistants as augmenting human work through faster, more accessible, and more scalable decision support, not replacing human oversight.
Which brand won the football tournament footwear battle
An analysis of footwear conversations around the 2026 World Cup compares how sponsorship, search demand, social attention, and sentiment shaped the competition between Adidas, Nike, Puma, New Balance, and other brands. Consumer Research tracked tournament-related mentions from June 11 to July 19 and paired them with search queries across the US, UK, and Canada. Adidas led tournament-specific footwear conversation with 63% of mentions, while Nike led broader brand discussion with 995k mentions, brand search momentum, and 78% positive sentiment. Nike’s cultural activations, including LISA’s “Rip The Script” campaign and Ronaldo’s Mercurial moments, generated shareable attention, whereas Adidas benefited from official sponsorship, the Trionda match ball, and player storytelling. Pink cleats became a cross-brand liability, attracting more than 5.5k mentions with nearly 70% negative sentiment, showing that visibility did not automatically produce favorable attention.
Top topics
- Architecture834
- Performance816
- LLMs660
- Machine Learning622
- Open Source563
- AI Agents553
- AI470
- Developer Experience448
Weekly digest
The best engineering reads, once a week.