Loading…
Browse by type
Explainer reads
Explainer posts on Yomu, summarised newest first.
- 1,318 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.
Explainer posts
Mixture of Experts (MoEs) in Transformers
Mixture of Experts (MoEs) in Transformers let a Transformer retain its backbone while replacing selected feed-forward layers with routed expert sub-networks, allowing total model capacity to exceed the active parameters used per token. The post explains why sparse models can reduce inference cost and memory pressure, then focuses on making them first-class in Transformers. Its weight-loading refactor converts independently serialized expert tensors into packed runtime layouts through composable WeightConverter operations and lazy materialization. The Experts Backend supplies eager, batched_mm, and grouped_mm execution strategies, while expert parallelism shards experts across devices and aggregates partial outputs. Training support uses grouped GEMM and custom Triton plus LoRA kernels, with reported improvements of approximately 12× speed, over 35% VRAM reduction, and roughly 6× longer context.
The generative recommender behind Shopify's commerce engine
Shopify describes a foundational generative recommender that predicts next products, and in some contexts ads, from buyer event sequences spanning searches, views, add-to-carts, favorites, and purchases. The system uses an autoregressive model with causal masking, time-aware attention combining RoPE-inspired rotary encoding with relative attention bias, and sampled softmax with shared and positive-aware hard negatives. Built on HSTU, it combines richer activity data, ensemble-focused training for incremental recall, and optimized CUDA kernels; the full training pipeline reached up to 7.3x the speed of a baseline implementation. In an online test of an August model version, relative shop orders rose 0.94%, high-quality click-through rate rose 5%, conversion rate rose 0.71%, and final served product recall at 2 rose 4.8%.
How Notion uses Custom Agents
Notion describes how Custom Agents are being used internally to automate repetitive questions, triage incoming work, and support communications, security, recruiting, and office operations. Agents are configured by describing a workflow and adding a trigger or schedule; they operate across Notion, Slack, Mail, Calendar, and other tools while using existing knowledge for context. Examples include a CX agent that searches the CX Knowledge Base and enablement docs, a Feedback Router that classifies Slack feedback and creates tasks, and Wilbur, which triages web issues and tags the right person. The company says these workflows make knowledge available beyond individual experts, reduce manual routing and response delays, and shift attention toward complex problems and creative work.
Detecting and preventing distillation attacks
Anthropic reports identifying industrial-scale distillation campaigns by DeepSeek, Moonshot, and MiniMax that generated more than 16 million Claude exchanges through approximately 24,000 fraudulent accounts. The campaigns used proxy services and coordinated account networks to extract agentic reasoning, tool use, coding, computer-use, and other capabilities, including chain-of-thought data and reinforcement-learning tasks. Anthropic attributed the operations with high confidence using IP correlation, request metadata, infrastructure indicators, and partner corroboration, while describing hydra cluster architectures that replace banned accounts and obscure patterns. The response combines classifiers, behavioral fingerprinting, coordinated-activity detection, intelligence sharing, stronger account verification, and product, API, and model-level countermeasures. The report argues illicitly distilled models may strip safeguards and undermine export controls, and says coordinated action across AI labs, cloud providers, and policymakers is required.
Two years of vector search at Notion: 10x scale, 1/10th cost
Notion describes two years of scaling vector search for AI Q&A, where semantic embeddings retrieve workspace and connected-tool content by meaning rather than exact keywords. The initial system combined Apache Spark batch indexing with Kafka-driven real-time updates, workspace-based sharding, and generation routing; rapid demand drove a 600x increase in daily onboarding capacity and cleared a waitlist of millions of workspaces by April 2024. A serverless migration cut vector database costs 50 percent from peak usage, while a later move to turbopuffer reduced search-engine spend 60 percent, lowered AWS EMR costs 35 percent, and improved p50 query latency from 70–100ms to 50–70ms. Further work uses xxHash-based page-state caching in DynamoDB to reprocess only changed spans and shifts embeddings generation from Spark and query embeddings away from a third-party API to Ray on Anyscale, with early rollout results pointing to a projected 90+ percent reduction in embeddings infrastructure costs.
Minions: Stripe’s one-shot, end-to-end coding agents—Part 2
Stripe’s minions are an unattended agentic coding flow whose pull requests are human-reviewed but contain no human-written code; more than 1,300 merge weekly. The system runs agents in AWS EC2-based devboxes that are prewarmed to become usable within about 10 seconds, providing parallel, predictable, and isolated environments. Its custom harness uses code-defined blueprints that combine deterministic nodes, such as linting and pushing changes, with agent loops for tasks such as implementation and fixing CI failures. Minions receive automated feedback through local linting, one full CI iteration, and a possible second repair attempt, while devbox isolation and security controls limit destructive actions; Stripe presents the approach as an extension of long-running investments in human developer productivity.
First Block with Gabe Pereyra, Co-founder and President of Harvey
The episode presents Harvey’s founding and growth through Gabe Pereyra’s account of building an AI platform for large law firms and their clients. Two roommates—Winston, a securities litigator, and Gabe, an AI researcher—identified an opportunity in legal. Harvey chose a general-purpose platform over a narrow workflow such as NDA review, aiming for a comprehensive AI associate capable of handling varied work. The discussion also covers enterprise accuracy, infrastructure over models, and the difficulty of turning early seven-figure deals into a repeatable sales machine, with investors helping on that learning curve. Harvey’s trajectory includes a seed check from OpenAI and a valuation reported at $8 billion, while the episode frames domain expertise and scalable go-to-market execution as central founder lessons.
How Stripe rolled out a consistent Cursor experience for 3,000 engineers
Stripe rolled out Cursor to more than 3,000 engineers while preserving its standards for developer experience and API uptime. The company preloaded Cursor on every machine, taught it in onboarding labs alongside VS Code and IntelliJ, and used Cursor Rules to supply codebase and coding-standard context, with teams able to add local rules. Adoption was reinforced by power users, while code review evolved to use LLMs to flag complex methods and risky files so reviewers could focus attention without sacrificing quality. An internal survey found developer sentiment at a five-year high, and MacVicar observed the largest productivity gains among longest-tenured engineers rather than junior developers. That finding has prompted exploration of letting new hires learn the codebase before receiving Cursor access, reinforcing MacVicar’s advice to start before every AI-workflow question is settled.
Everyone wants AI. That isn't enough.
Survey data from more than 3,000 global knowledge workers and 85-plus interviews examines why strong demand for workplace AI has not translated into broad adoption. Although 93 percent call AI essential to modern productivity software and 80 percent are likely to switch to a platform offering it, only 20 percent say AI is deeply integrated into their company’s tools. Respondents prioritize recurring workflows such as automated project updates, continuously current knowledge hubs, and goal-based templates, while decision-makers identify integration as the leading implementation challenge alongside security and data quality, unclear strategy, resource constraints, and compatibility. The proposed pattern is to place AI inside existing work, illustrated by Faire’s reported results with Notion AI: smoother meetings for 81 percent of surveyed staff and workflows running 30 percent faster.
CX Score: How we built a metric support leaders can defend
CX Score was built to give support leaders a defensible view of customer experience across conversations, addressing whether an automated metric can be trusted. The team grounded its criteria in how experienced support professionals assess strong, mixed, and poor interactions, then created a dataset of thousands of conversations reviewed by experts, with two reviewers where possible and disagreement resolution. It operates as a production system using specialized prompt engineering, support-focused model configuration, neutral defaults for ambiguous evidence, and multi-stage evaluation across LLM models. Precision, recall, F1, regression, and bias checks set a required F1 above 0.8, while field testing with real customers examined varied tickets, languages, agent types, and conversation lengths. The stated conclusion is that CX Score is designed to be inspected, explained, and defended rather than treated as a black-box output.
How European companies build AI-ready knowledge foundations
European organisations face a knowledge paradox: 97% of surveyed business leaders say knowledge management is critical, but only 28% report having a coherent strategy, with poor practices costing up to €16,000 per employee annually. Interviews with leaders in France, Germany and the UK connect scattered information across people and tools to slower progress, weaker decision confidence, repetitive work and lost institutional knowledge. The post argues that centralised knowledge is a prerequisite for effective AI, noting that 56% of organisations already use AI for knowledge management and 29% are piloting it, while incomplete documentation can undermine deployments. Recommended practices include explaining the personal benefit of documentation, designing systems that pull participation rather than enforce compliance, and treating knowledge as a shared organisational asset. Examples include Nelly, Above, UnternehmerTUM and EGYM using simpler documentation workflows and shared spaces to reduce silos and keep information current.
OpenEnv in Practice: Evaluating Tool-Using Agents in Real-World Environments
OpenEnv is an open-source framework from Meta and Hugging Face for evaluating AI agents against real systems rather than simulations, with stateful environments that connect to tools and APIs. Turing's production-grade Calendar Gym applies this model to calendar operations, combining access control lists, partial visibility, temporal reasoning, and multi-step workflows in isolated sessions. Agents often handled individual actions but became less reliable on longer or ambiguous tasks: success was close to 90% with explicit calendar identifiers and roughly 40% when requests used natural-language descriptions. More than half of failures involved malformed arguments or incorrect action ordering despite selecting the correct tool, while structured validation, permission, and datetime errors provided concrete recovery signals.
Minions: Stripe’s one-shot, end-to-end coding agents
Stripe’s Minions are homegrown, fully unattended coding agents designed to complete tasks in a single run, beginning with an invocation such as a Slack thread and ending with a pull request ready for human review. More than 1,000 pull requests merged weekly at Stripe are minion-produced and contain no human-written code, although engineers review them and can provide follow-up instructions. Each run uses an isolated, pre-warmed devbox and a customized fork of Block’s coding agent goose, combining agent loops with deterministic operations for version control, linting, and testing. Minions use curated MCP tools, including Stripe’s Toolshed with more than 400 tools, to retrieve internal context before and during work. Local checks and selective CI provide feedback, with no more than two CI rounds, balancing correction against token, compute, and time costs.
NVIDIA commits 3x more code across 30,000 developers with Cursor
NVIDIA has deployed Cursor across its software development lifecycle, with more than 30,000 developers using it daily and committing over three times as much code as before. NVIDIA’s long-lived, interdependent codebases and varied technology stacks made it difficult to understand downstream effects and keep pace with rapid change. Cursor maps and semantically reasons over large codebases, retrieving relevant context, while custom rules and MCP servers extend it from code generation into reviews, testing, debugging, issue-driven fixes, CI troubleshooting, and git-flow automation. The company also uses it to help new hires learn unfamiliar systems and experienced engineers work across programming languages and stack areas. NVIDIA reports that bug rates have remained flat despite higher coding velocity, while code-style consistency has improved.
Engineering at scale, with purpose.
Senior software engineer Brett Shouse outlines site reliability engineering initiatives and infrastructure modernization efforts underway at Thumbtack. The engineering organization is currently leading an operational project to migrate multiple self-hosted observability services onto a single, unified SaaS platform. This architectural transition consolidates application logs, distributed traces, and system metrics into one accessible view tailored for engineers, customer support staff, and company executives. Moving away from legacy self-hosted monitoring systems reduces systems administration overhead, mitigates alert fatigue, and lowers direct operational infrastructure costs. Concurrently, the reliability team tackles technical debt accumulated from rapid organizational growth by establishing structured incident response processes and automating repetitive operational toil.
Global Consumer Insights Ahead of 2026 Ramadan
Ramadan reshapes daily routines, online activity, spending, dining, and hospitality, making it a major consumer and cultural moment across Muslim regions. Drawing on nearly 7 million English- and Arabic-language online conversations from 2025, the analysis tracks shopping, food, home aesthetics, hospitality, and regional posting patterns. Ramadan mentions rose 76% in the three months before the 2026 holiday year over year, while fragrance mentions increased 33%, Iftar-party mentions approached 900,000, recipes rose 144%, and Iftar or Suhoor packages and ticketed events rose 145% in 2025. Joy accounted for 57% of emotion-categorized Ramadan mentions, and conversations centered on togetherness, spirituality, traditions, and generosity. Afternoon activity peaks between 2 pm and 4 pm across the analyzed regions, but brands are advised to use their own regional and platform analytics and meet audiences through authentic participation rather than promotion.
Towards self-driving codebases
The post describes research into scaling long-running autonomous coding agents, using a web browser as a complex benchmark and running thousands of agents continuously for one week. An initial single-agent plan failed because the model lost track, stopped prematurely, and stalled on complex details; shared-state coordination then introduced locking contention, while more structured planner, executor, worker, and judge roles improved ownership and throughput. The team built a Rust-based harness on one large Linux VM, logged agent messages and tool activity for replay and analysis, and iterated through designs that eventually favored a more dynamic system with freshness mechanisms and fewer overloaded roles. The research concludes that architecture, prompts, observability, explicit throughput trade-offs, and anti-fragile recovery matter, reporting linear scaling of token throughput with minimal harness overhead while humans retained taste, judgment, and direction.
Defining customer support in 2025: Why it's key to your long
Customer support has shifted from a perceived cost to a differentiator as expectations rise and companies focus on retention, advocacy, and product feedback. Modern support aims to resolve issues, troubleshoot problems, and help customers get more value through personal, contextual, knowledgeable, and proactive interactions. The post argues that human+AI partnerships can improve speed and efficiency: AI agent Fin handles common questions in more than 45 languages, while human teammates address complex queries empathetically. It also explains why ticketing became necessary for tracking asynchronous work, internal ownership, and customer updates within an all-in-one support platform. Examples from Hospitable and Rebag report faster responses, reduced response times, and higher NPS, while the broader guidance emphasizes clear expectations, robust processes, escalation paths, and a clear route to human help.
The Future of the Global Open-Source AI Ecosystem: From DeepSeek to AI+
This third and final article in a series examines how Chinese AI organizations are shaping an open-source ecosystem after DeepSeek R1's January 2025 release. It argues that open sharing of models, papers, deployment infrastructure, and techniques has become the dominant near-term approach, with organizations ranging from Alibaba, Tencent, ByteDance, and Baidu to startups and application companies contributing at different layers. Qwen illustrates the ecosystem strategy: by mid-2025, more than 113,000 Hugging Face models used Qwen as a base, while organizations also connected models with chips, cloud platforms, toolchains, evaluation, and deployment systems. The article places this shift on infrastructure built since 2017, including the East Data, West Compute plan, and says R1 catalyzed wider use of existing compute, energy, and data capacity. It concludes that open-source models are becoming reusable, composable components in industrial workflows and localized AI systems as China moves from DeepSeek toward AI+.
Cursor at Grab: Adoption and impact
Grab introduced Cursor in late 2024 as one of several AI coding assistants and reports rapid adoption across engineering and non-technical teams. Monthly active usage reached 98% of Tech Grabbers, weekly usage about 75%, and Cursor suggestions were accepted around 50% of the time, compared with a cited industry average of 30%. About half of users engage on at least 10 consecutive days monthly, while more than a third of merge requests incorporate Cursor; common uses include test generation, refactoring, code navigation, bug fixing, and routine automation. Grab built monorepo indexing and distributed preconfigured rules for environment-specific suggestions, while a fixed-effects regression effort is examining productivity; early findings report a statistically robust dose-response relationship, with gains increasing alongside usage intensity.
Top topics
- Architecture830
- Performance812
- LLMs648
- Machine Learning617
- Open Source561
- AI Agents542
- AI454
- Developer Experience446
Weekly digest
The best engineering reads, once a week.