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
Transformation in action: How AI is evolving support careers
AI is changing customer service operations by shifting human agents away from queue-level work such as triage and FAQs toward training systems, complex escalations, and strategic improvement. In research from The 2026 Customer Service Transformation Report, 45% of teams updated job descriptions with AI-related responsibilities, 40% said agents focus more on training AI systems, and 27% primarily handle complex escalations and edge cases. Interviews with 166 customers found nearly 95% reporting meaningful workflow changes, while 83% saw roles become more strategic and supervisory; mature teams also reported less time handling support volume. New positions including AI knowledge managers and conversation designers are emerging, while existing staff can move into AI-focused work such as optimizing knowledge, monitoring outputs, and improving customer journeys. The source recommends transparent communication, agent participation, dedicated time outside inbox, and metrics such as resolution rate as teams build operating models with AI at their core.
2,000 robots walk into a shop: Simulated A/B testing (2026) - Shopify
Shopify’s SimGym addresses the difficulty of evaluating storefront changes when merchants lack enough traffic for conventional A/B tests, by simulating shopping sessions with LLM-guided robots. Each robot runs in a cloud Chromium browser with a persona, budget, and intent, while a counterpart tests the alternate theme; the system uses real page representations rather than a mocked DOM. To support bursty, sequential workloads with 89K–127K accumulated tokens per session, Shopify self-hosted gpt-oss-120b on 48 NVIDIA B200 GPUs, combining MXFP4 expert quantization, FP8 KV-cache, FlashInfer kernels, async scheduling, and guided decoding. Production measurements report 80K to 88K tokens/second per B200 from async scheduling, a 57% reduction in HTTP/gRPC overhead, and MIG experiments yielding about 20% lower latency and 12% higher daily throughput without quality regression; speculative decoding is validated but queued.
The third era of AI software development
The post describes a third era of AI-assisted software development, in which autonomous agents handle larger tasks over longer timescales while developers define problems and review results. It traces a progression from Tab autocomplete, to synchronous prompt-and-response agents, and then cloud-based agents running on separate virtual machines. These agents can iterate and test for hours, returning logs, video recordings, and live previews that make parallel work and artifact-based review practical. Cursor reports that 35% of its internally merged pull requests are created by agents operating autonomously in cloud VMs, and that agent users now outnumber Tab users two-to-one after usage grew more than 15x in a year. The post says broader adoption still depends on reliable tests, environments, tools, and context, while developers shift toward decomposition, review, and feedback rather than line-by-line guidance.
Using LLMs to amplify human labeling and improve Dash search relevance
Dropbox Dash uses large language models (LLMs) to amplify human labeling efforts for training its search relevance and ranking models. By validating and optimizing LLM evaluators against a small set of human-labeled internal data, Dropbox creates massive, high-quality training datasets offline for production rankers like XGBoost without incurring high latency or latency costs at query time.
Closing the code review loop with Bugbot Autofix
Bugbot Autofix extends the code review agent so it can find issues in pull requests and automatically propose fixes. It spawns cloud agents in independent virtual machines that test software, and more than 35% of its changes are merged into the base PR. Now out of beta and available to all Bugbot users, the feature adds proposed fixes to the pull requests Bugbot reviews once enabled. Bugbot’s average issues identified per run has nearly doubled over six months, while the user resolution rate rose from 52% to 76%, indicating more bugs are caught and fewer false positives are flagged. Future work includes configurable automations, self-verification, deeper research, and continuous codebase scanning.
New book: Intercom on Jobs
Intercom launched its fourth book, Intercom on Jobs-to-be-Done, to document lessons from applying the Jobs-to-be-Done framework to product, design, and marketing for software. The book addresses a gap: although the framework is nearly 30 years old and often discussed through physical products, few materials cover its application to software companies. Intercom says the framework helped it understand the needs customers addressed with its product and improve their experience. Rather than offering a prescriptive, step-by-step method, the book presents hard-learned lessons intended to be timeless and universal, including customer switching, real competitors, product scope, exact jobs, Job Stories versus personas, and software research.
Dario Amodei on the Department of War discussions
Dario Amodei says Anthropic considers AI deployment for US and allied democratic defense existentially important and has deployed Claude across the Department of War and other national security agencies, including classified networks and National Laboratories. The company says it supports intelligence analysis, modeling and simulation, operational planning, and cyber operations, while restricting Claude’s use by firms linked to the Chinese Communist Party and advocating chip export controls. Anthropic maintains two safeguards: mass domestic surveillance is incompatible with democratic values, and current frontier systems are not reliable enough for fully autonomous weapons without proper oversight. The Department has stated it demands “any lawful use” and threatens offboarding, a “supply chain risk” designation, and Defense Production Act action; Anthropic says it will not remove the safeguards, while offering continued service or transition support.
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.
Announcing major updates to Procedures and Simulations: Enabling Fin to handle complex queries
Intercom announces 12 updates to Fin’s Procedures and Simulations, the mechanisms it presents for handling complex customer queries while keeping teams in control. Procedures combine natural-language instructions, deterministic controls such as conditional steps, data connectors, code snippets, checkpoints, and fully agentic behavior; new features include AI drafting, reusable Sub-procedures, explicit switching rules, targeted knowledge-hub references, internal handoff notes, automatic switching, and structured extraction from PDFs and images. The Procedures UI also adds improved authoring and reporting on triggers, resolutions, and handoffs. Simulations support full multi-turn conversations, image uploads, clearer visibility into Fin’s reasoning, and AI-generated tests stored in a central library that can be rerun after changes. Procedures and Simulations are available now, with the stated goal of helping teams train, test, deploy, analyze, and refine complex workflows before and after launch.
Innovating with MongoDB | Customer Successes, February 2026
Modern artificial intelligence startups frequently face operational bottlenecks when deploying autonomous agents and dynamic interfaces on rigid relational schemas or restricted database engines. To address these limitations, platforms including Modelence, Thesys, Emergent Labs, and Heidi integrated MongoDB Atlas as their operational data foundation. Emergent Labs eliminated recursive schema migration failures previously encountered with PostgreSQL by adopting document structures that match native JSON generation, supporting millions of deployed applications. Meanwhile, Heidi migrated from Amazon DocumentDB to utilize MongoDB Vector Search for retrieval-augmented generation pipelines under a single unified API without requiring external search databases. These architectural transitions streamlined continuous agent iterations, eliminated operational scaling downtime, and accelerated deployment cycles across diverse production environments.
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.
Introducing Custom Agents
Notion has made Custom Agents available to Business and Enterprise plans as autonomous AI teammates for recurring work across Notion, Slack, Mail, Calendar, Figma, Linear, and custom MCP servers. Users describe a job and set triggers, while agents can answer questions, route tasks, compile updates, and run workflows across connected tools; setup includes an agent-written instruction set and tool configuration. The announcement describes enterprise controls including permissions, version control, usage analytics, logged runs, reversible changes, alerts, auto-pause, and administrative connection controls, while noting prompt-injection risks and planned guardrails. Custom Agents are free through 3 May 2026, then consume Notion credits based on work, with the seat price and other listed AI features unchanged; early testers have built over 21,000 agents.
Cursor agents can now control their own computers
Cursor describes cloud agents running in isolated virtual machines with full development environments, letting them use and test the software they create rather than stopping at code generation. Available through the web, mobile, desktop, Slack, and GitHub, the agents onboard onto codebases, produce merge-ready pull requests, and attach videos, screenshots, and logs as validation artifacts. Each agent receives an isolated VM for parallel work and can interact directly with browsers, desktop applications, and other tools; users can control its remote desktop and edit without checking out the branch locally. Examples include implementing and verifying Marketplace source links, reproducing a clipboard-exfiltration vulnerability, testing lint-label changes in Cursor, and walking through cursor.com/docs. Cursor reports that more than 30% of its merged PRs are created by autonomous cloud-sandbox agents, while its next goals include coordinating agents, learning from past runs, and eventually managing production changes end to end.
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.
We proactively fixed ~100 security issues in 6 days with 0 humans
Ramp Security Engineering describes a six-day experiment in which home-grown agents found, validated, and fixed nearly 100 latent backend security issues, including some high-severity findings, without human involvement until pull-request review. The issues had not been uncovered by penetration testing, bug bounty testing, static analysis, or trials of 10+ code-scanning vendors, and all were patched within a week of discovery. The pipeline used specialized vulnerability detectors, adversarial manager agents that rejected 40% of initial proposals in sample testing, integration tests for validation, and an internal coding agent that generated patches. Live interactive validation struggled with complex preconditions, so test-based validation became the main approach; a human team member still reviewed and landed each pull request.
GGML and llama.cpp join HF to ensure the long-term progress of Local AI
Hugging Face announces that GGML, the creators of llama.cpp, are joining the company to support the ggml and llama.cpp community as local AI advances. Georgi Gerganov and his team will continue dedicating 100% of their time to maintaining llama.cpp, retaining autonomy and leadership over its technical direction and community, while the project remains fully open-source and community driven. The technical focus is tighter integration between llama.cpp’s local-inference stack and Transformers’ model definitions, aiming to make shipping new models from Transformers’ “source of truth” nearly single-click, alongside better packaging and user experience for ggml-based software. The longer-term goal is an efficient inference stack that runs on devices and helps make open-source superintelligence accessible, with llama.cpp becoming ubiquitous and easier for casual users to deploy and access.
Train AI models with Unsloth and Hugging Face Jobs for FREE
The post presents a workflow for fine-tuning the LiquidAI/LFM2.5-1.2B-Instruct small language model with Unsloth on Hugging Face Jobs, including optional use of Claude Code, Codex, or another coding agent. It explains that Unsloth is described as providing about 2x faster training and 60% lower VRAM use than standard methods, while the model runs under 1GB of memory and is optimized for on-device deployment. The tutorial covers claiming free Hugging Face Jobs credits through the Unsloth Jobs Explorers organization, installing the hf CLI or a model-training skill, and submitting a job with a dataset, GPU flavor, timeout, and output repository. An agent can generate a UV training script using 4-bit loading and LoRA, submit the managed cloud-GPU job, report Trackio monitoring details, and push the trained model to the Hugging Face Hub.
Scaling Localization with AI at Lyft
Lyft re-architected its batch localization pipeline to combine artificial intelligence with professional linguist oversight. The previous workflow relied exclusively on human translation, which resulted in multi-day turnarounds and linear cost scaling that hindered rapid market expansions. Under the new architecture, a dual-path system generates three candidate translations using a fast generation model, evaluates them against defined rubrics using a reasoning-focused model, and immediately releases approved drafts. Linguists review the drafts asynchronously in the Smartling translation management system to establish the final system of record. This pipeline reduced delivery latency from days to minutes while achieving a state where 95% of machine-generated translations require no significant changes during human review.
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.
Top topics
- Architecture834
- Performance816
- LLMs660
- Machine Learning622
- Open Source563
- AI Agents553
- AI470
- Developer Experience448
Weekly digest
The best engineering reads, once a week.