Loading…
Latest reads
Every post, summarised
Grounded summaries of engineering blogs, newest first. Each links to the original.
- 4,668 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
Hollywood demos vs. real
AI demonstrations fall into two categories: polished “Hollywood demos” and raw “real-world demos,” with the latter showing how customer-service AI agents perform under messy, complex interactions rather than edited presentation. Voice provides an especially demanding test because timing, tone, turn-taking, latency, interruptions, and real-time adaptation all shape the experience. At Pioneer, Paul Adams called Fin Voice live; within 90 seconds, it verified identity, retrieved account data, handled an interruption, offered options, completed the workflow, and sent a follow-up email. The demonstration exposed brief pauses for backend work, natural conversational flow, tone changes, and unscripted behavior, while later updates added voice customization, escalation guidance, deployment controls, integrations, support for 28 languages, and roughly 30–40% lower latency since launch.
Heroku Support for .NET 10 LTS: What Developers Need to Know
Heroku now supports building and running .NET 10 applications, including ASP.NET Core 10, C# 14, and F# 10, as the new three-year Long Term Support release becomes available. Its .NET buildpack detects single-file C# applications, installs the latest SDK, builds and publishes them, and configures web apps without project or solution files; it also supports both .sln and .slnx formats. Heroku CI continues to run dotnet test by default, with the .NET 10 SDK integrating the Microsoft Testing Platform into that command. .NET 8 and .NET 9 remain supported with updates but reach End of Support on November 10, 2026, while apps targeting net10.0 RC will use stable .NET 10 on their next push and can pin SDK versions through global.json.
The hidden cost of disconnected knowledge
Disconnected knowledge leaves important decisions in Slack threads, meetings, and Jira comments rather than the wiki, separating information from the work that uses it. The post frames this fragmentation as three compounding costs: switching between tools, paying for tools employees do not adopt, and limiting AI because it cannot access conversations, decisions, projects, and tasks. It proposes shrinking the distance between knowledge and action through a connected workspace, while keeping tools such as Slack, Figma, and GitHub searchable through AI. In this model, documentation is created during work, remains current, and links specifications, roadmaps, OKRs, and projects. A framework evaluates organizations as Fragmented, Consolidating, or Integrated across connected, alive, accessible, intelligent, and adopted dimensions, with examples from Ramp, OpenAI, Faire, and Cursor.
Improving agent with semantic search
Cursor describes semantic search as a complement to grep for helping coding agents locate relevant code in large codebases. Its evaluation used Cursor Context Bench across widely used models, comparing tool sets with and without semantic search, and found an average 12.5% accuracy improvement, with gains ranging from 6.5% to 23.5%. In an online A/B test, semantic search increased code retention by 0.3% overall and 2.6% on codebases with at least 1,000 files, while its absence produced a 2.2% increase in dissatisfied follow-up requests. The retrieval system uses a custom embedding model trained from agent session traces, with an LLM ranking helpful content and training aligning embedding similarity scores to those rankings. The conclusion is that combining semantic search with grep produces the best outcomes, especially for large codebases.
Grab's Mac Cloud Exit supercharges macOS CI/CD
Grab moved its macOS CI/CD infrastructure from a US cloud vendor to a colocation cluster in Malaysia, closer to its Git servers in Southeast Asia. The change followed growth from one office Mac Pro to more than 250 Mac minis, rising demand, job queues, and cloud costs that made ownership or colocation more attractive at scale. Grab selected bare-metal hardware after comparing cloud, colocation, and on-premises options, then migrated progressively while monitoring failure rates, performance, and reliability; the later infrastructure also uses zero-touch provisioning to reduce manual setup as the cluster grows. The migration produced projected savings of more than $2.4 million over three years and performance improvements of 20–40%, while the team chose allocated-kilowatt power commitments and retained bare metal because macOS virtualization’s performance or stability trade-offs did not justify it.
You Should Write An Agent
Thomas Ptacek argues that people should build an LLM agent before forming strong opinions about the technology, because implementation reveals how simple and surprising the underlying system can be. The example starts with a Python loop around the OpenAI Responses API, storing user and assistant messages in a context list to simulate conversation with a stateless model. Adding a JSON-defined ping function and a tool-calling loop lets the model choose and execute multiple network checks, demonstrating behavior not explicitly hard-coded in the loop. The discussion extends this pattern to coding agents, contexts persisted in SQLite, sub-agents, and design choices involving nondeterminism, ground truth, interchange formats, and token costs. It concludes that these remain open software-engineering problems whose practical behavior can be understood by building and iterating on agents.
Upgrade to Puma 7 and Unlock the Power of Fair Scheduled Keep-alive
Puma 6 and earlier could let a keep-alive connection “cut the line” when all threads were busy, allowing up to ten requests from that connection to run consecutively while previously accepted work waited. The behavior came from reusing a “fast inline” optimization intended for HTTP pipelining, where requests arrive together, for keep-alive requests that can arrive later. Puma 7 changes queueing so pipelined and keep-alive requests return to the back of the queue, and adjusts wait_until_not_full so interleaving is possible; its accept loop now slows proportionally with load instead of stopping completely. The fix does not add capacity, but distributes overload delay more fairly across requests. The post recommends upgrading to Puma 7.1.0 or later, especially for applications that had disabled keep-alives after latency spikes.
Building an Enterprise Data Warehouse on Heroku: From Complex ETL to Seamless Salesforce Integration
The article describes Heroku’s effort to unify Salesforce and data from more than 20 applications in a real-time analytics warehouse after traditional ETL created API, freshness, performance, and maintenance problems. Its architecture uses Heroku Postgres as an operational warehouse and staging layer, AWS Redshift for historical analytics, Heroku Connect for direct Salesforce replication, follower databases for read-only source access, and Apache Airflow for orchestration. Template-driven Python jobs and 200+ JSON definitions support incremental, bulk, and date-range loads, while 30+ DAGs and monitoring across 12 regional Heroku Connect instances provide automated retries and consistency. Reported outcomes include over 10 TB processed monthly, sub-minute freshness, 99.99% pipeline uptime, 60% lower integration costs, and 40% better customer analysis accuracy. Tableau and Dataclips provide real-time, historical, and ad hoc reporting, while planned or newly described capabilities include Postgres Advanced, expanded Connect synchronization, and zero-copy Data Cloud ingestion in beta.
Shipping is your company's heartbeat
Software becomes valuable only when it reaches customers, because deployment starts the feedback loop that tests assumptions and enables the next improvement. The post argues that although software deployment can now take seconds or minutes, teams often retain habits from an era of infrequent, costly releases, making shipping itself a source of risk and delay. It uses time to production (TTP) as a technical and organizational lens: a one-minute fix can be tested, reviewed, merged, deployed, and communicated immediately, while hours, days, or weeks encourage waiting, batching, deploy windows, larger releases, and growing bug backlogs. Continuous shipping is presented as a company-wide practice that improves feedback, decision-making, customer responsiveness, momentum, motivation, and hiring, while requiring ongoing effort to reduce TTP and treat shipability as a design concern.
Learn How to Lower Heroku Dyno Latency through Persistent Connections (Keep-alive)
Repeatedly opening TCP connections between the Heroku Router and an application dyno can impose TCP slow start on every request, creating a latency penalty. The post explains how HTTP/1.0 tied each request to a new connection, while HTTP/1.1 keep-alive connections reuse an open connection and amortize setup overhead; unlike pipelining, they do not require requests to be prepared in advance. Heroku Router 2.0 uses keep-alive connections to dynos by default, avoiding repeated slow-start costs on repeat requests and exposing more available bandwidth and lower latency. Ruby applications that previously disabled keep-alives are directed to upgrade to Puma 7.1.0 or newer, then update the bundle, inspect settings, and deploy. Keep-alives still depend on both Puma and router configuration, and the post notes that the router also supports HTTP/2 features.
How we built a custom vision LLM to improve document processing at Grab
Grab developed a specialized Vision LLM for extracting text and key information from Southeast Asian identity cards, driver’s licenses, and registration certificates, where varied languages and layouts challenged traditional OCR and existing models. It selected Qwen2-VL 2B for its efficient size, SEA language tokenizer coverage, and dynamic native-resolution image handling, while synthetic Common Crawl text images and Documint supplied reviewed training labels for detection, orientation, OCR, and KIE. LoRA fine-tuning improved Latin-script documents but struggled with Thai, Vietnamese, and dense unstructured layouts, leading to two-stage full-parameter training with synthetic SEA text images followed by task-specific document fine-tuning. The fully fine-tuned 2B model raised Thai accuracy by 70 percentage points and Vietnamese accuracy by 40 points over baseline, while a custom roughly 1B-parameter model combining Qwen2-VL 2B’s vision encoder with Qwen2.5 0.5B’s decoder stayed within a 3-point accuracy gap across most document types and achieved lower latency.
MongoDB Announces Leadership Transition
Dev Ittycheria announced his decision to retire as MongoDB President and Chief Executive Officer after eleven years of leading the company. Chirantan Desai will become the new Chief Executive Officer effective November 10, 2025. The transition follows a succession planning process initiated earlier this year when Ittycheria concluded he could not commit to an additional five-year term requested by the Board. Desai previously scaled ServiceNow organically from over $1 billion to more than $10 billion in revenue and served as President of Product and Engineering at Cloudflare. Ittycheria will remain on MongoDB's Board to support Desai through the transition into the company's next phase of growth.
Introducing Cursor for Enterprise
Cursor is introducing enterprise features intended to make its coding agents more capable, transparent, and secure for organizations already using the product. Hooks expose the agent loop to custom scripts that can log actions and tool calls, enforce compliance or command allowlists, scrub sensitive material, connect external systems, and trigger automations; they can be distributed through MDM or Cursor’s cloud option. Team Rules provide shared context and can recommend or require standards such as API schemas, conventions, and workflows, while upgraded analytics report activity, adoption, AI lines of code, directory-group filters, and exportable data with updates every two minutes. The release also adds an Audit Log covering 19 event types and Sandbox Mode, which restricts network and file access by default, with administrators controlling availability and team-wide Git and network access.
How to build a Content Design Agent
An overloaded solo content designer built VERBI, an AI content design agent, to scale standardized work without replacing human judgment or filling products with generic copy. The agent runs in Glean, using a curated collection of Intercom’s style guide, tone of voice, product and monetization glossaries, messaging guides, research, and official websites as retrieval-augmented generation sources. The system prompt defines its role, use cases, tone, and component rules, while a Glean Indexing API connection makes Storybook’s Surge design-system guidance available as a live source. Additional instructions tell VERBI to challenge users, avoid guessing when sources lack an answer, and keep responses concise. After 700+ triggers, the reported benefits are consistent content rules, faster drafts and iteration, fewer repetitive questions and errors, and more time for higher-impact design review.
Heroku Connect: Faster, More Reliable Data Sync with Salesforce CDC
Heroku Connect customers needing current Salesforce data for applications such as custom quoting have faced stale records because legacy synchronization could lag, with Opportunity Line Items among unsupported objects for Accelerated Polling. Heroku is rolling out support for Salesforce Change Data Capture (CDC), allowing Accelerated Polling to listen for changes instead of relying only on the Salesforce Streaming API. That change adds more than 170 Salesforce standard objects to Accelerated Polling, including Opportunity Line Items, Pricebook2, Asset, and OpportunityLineItem, so critical data can synchronize more quickly. To enable it, a Salesforce administrator turns on CDC for selected objects, after which the corresponding Heroku Connect mappings expose the “Uses Accelerated Polling” option.
Headlines from Pioneer 2025: Fin 3, the vision for a unified Customer Agent, and what’s next for customer experience
Pioneer 2025 presented Intercom’s latest customer-service AI plans, centered on Fin 3 and a longer-term vision for a single Customer Agent covering the entire customer experience. Fin 3 adds Procedures for complex, policy-controlled resolutions, Simulations for pre-release testing, Slack and Discord deployment, improved Voice, and Insights features such as CX Score Reasons, Topic Trends, and AI-powered Suggestions. The proposed Customer Agent would combine specialized roles, user-defined goals, persistent memory, business knowledge, and interoperability across tools, systems, and channels. Customer examples described phased rollouts, rapid feedback, broad scaling, and support-team involvement, while Anthropic reported two specialists enabling nearly 50 full-time employees’ worth of coverage in its volume. The event framed AI as a force for converging products, roles, and organizational structures, with the Customer Agent future described as arriving quickly.
What’s new with Fin 3: The best AI Agent for complex queries across every channel
Fin 3 is presented as an AI Agent for customer service focused on increasing automation rate by handling complex queries end-to-end across more channels. Procedures combine natural-language instructions with data connectors, conditional steps, code snippets, and approval checkpoints, allowing teams to encode standard operating procedures with varying levels of control. Simulations provide multi-turn testing for different customer segments and scenarios, while Fin Voice, Slack, and Discord expand deployment options. Insights upgrades add CX Score Reasons, Topic Trends, and AI-powered Suggestions that surface quality changes, topic shifts, and content improvements. The announcement says Fin 3’s broader goal is to increase the share of overall support workload it handles, beyond a resolution-rate metric that had reached 66% on average across more than 6,000 customers.
Preparing for the Customer Agent future
At Pioneer 2025, Eoghan McCabe presented a plan for Fin to become a unified Customer Agent spanning the entire customer experience, from first contact through purchase, support, success, and upsell. The proposal responds to the risk that separate AI Agents for sales, marketing, success, and service will use conflicting priorities, disconnected information, and inconsistent interactions. Fin is intended to evolve from a task-based system into an agentic system with fluid roles, user-defined goals, persistent customer memory, comprehensive business knowledge, and interoperability across tools, systems, and channels. The post argues that customer service leaders are positioned to guide this wider AI rollout because they have experience, data, and operational insight from early AI adoption. Early activity includes WHOOP using Fin for sales conversations, where it is resolving 84%, while new capabilities are expected to roll out over the coming months.
Cloud Agents
Cursor’s Cloud Agents let developers run multiple agents remotely without keeping a laptop connected to the internet. Agents can be launched and managed from the Cursor editor, cursor.com/agents, Slack, Linear, and GitHub, supporting background bug fixes, small todos, and larger feature work. For difficult bugs, the team commonly has multiple models attempt the same problem and selects the best result; for complex features, a locally developed plan can be sent to a cloud agent for implementation, allowing work to continue on another task. Cursor also revamped its GPT-5 Codex agent harness for longer time horizons in the cloud, while plan mode supports sending implementation plans remotely. The workflow is available today in Cursor editor and Cursor Web, as well as the listed integrations.
Aligning to What? Rethinking Agent Generalization in MiniMax M2
MiniMax M2 addresses the performance gap between agent benchmark scores and real-world deployment across diverse scaffolding frameworks and environments. While initial alignment efforts relied on scaling tool variety, models continued to fail when subjected to slight variations in prompts, scaffolding, and tool responses. To resolve this instability, the team introduced interleaved thinking, which enables models to execute internal reasoning at arbitrary steps throughout long-horizon tasks to adapt dynamically to external tool outputs. Additionally, developers constructed a data pipeline designed for full-trajectory generalization by introducing perturbations across system prompts, user objectives, environments, and tool feedback. In internal evaluations, MiniMax M2 maintained robust tool-calling and instruction-following performance when tested against unfamiliar, cold-start agent scaffolding frameworks.
Top topics
- Architecture834
- Performance815
- LLMs659
- Machine Learning621
- Open Source563
- AI Agents553
- AI469
- Developer Experience448
Weekly digest
The best engineering reads, once a week.