Kevin Ho
美国 纽约 纽约
600 位关注者
500+ 位好友
查看与Kevin的共同好友
Kevin可以把您引荐给Stripe的 10+ 位会员
查看与Kevin的共同好友
动态
600 位关注者
-
Kevin Ho 转发了此动态Kevin Ho 转发了此动态AI agents are not going to replace software engineers yet, at least not when building Stripe integrations. The critical gap between our AI reality today and production-ready reality: we sorely lack a stable and realistic validation layer, especially when it comes to financial infrastructure where we require 100% accuracy. So to measure how AI is progressing on that last mile of value, Kevin Ho and I built a set of robust, replicable software engineering environments to benchmark models on their ability to build complete Stripe integrations. They mirror realistic integration scenarios: huge codebases, data migrations, breaking version changes, server and client-side API integrations, and even browser use. The range of results were extremely surprising, and even more exciting is how we now have an experimentation test bed to demonstrably improve how models integrate with Stripe. We’ve open-sourced the full benchmark suite in our AI toolkit: https://lnkd.in/em-aEwMF. Read the blog post: https://lnkd.in/eXAVSHRq.Can AI agents build real Stripe integrations? We built a benchmark to find outCan AI agents build real Stripe integrations? We built a benchmark to find out
-
Kevin Ho 分享了此动态Kevin Ho 分享了此动态Hitting my one month at Stripe and I have to say I am so impressed by: • The collaborative culture and empathy fellow Stripes lead with 🤗 • The awesome processes laid out and the greenfield to keep building and innovating🚀 • The value Stripe has for recruiting (my hiring managers are so passionate about building diverse talented teams, that they spend 30+ minutes a day sourcing on LinkedIn 🙌) • Also the swag is pretty great too 😎
-
Kevin Ho 分享了此动态Kevin Ho 分享了此动态Stripe Connect is growing quickly!🚀 Do you want to work with passionate leaders who value diversity and professional growth?🙋🏻♀️ Tushar Jain, Ilker Temir, Mario T., Ali Siddiqui, Brian Pratt and Sid Gidwani are looking for inspiring Engineering Managers who can grow and lead impactful teams! Want to learn more? Let's chat!Engineering Manager, Connect - Stripe: Job OpeningsEngineering Manager, Connect - Stripe: Job Openings
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态Thank you Alex Hernandez ☁️ Maria (Jacobs) Mann Alexandra Lersch and the team at Salesforce for an amazing Dreamforce! It was a great experience being able to share our story with Amin Bouabdellah and Stephanie Seville on the future of Salesforce in Claude.
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态At Stripe, interns elevate our ambitions every year. In the 2010s, one intern turned our laptop-based deployment scripts into our first centralized deployment system. Another built our original, homegrown status page. Another laid the foundation for our API permissions framework and rewrote API authentication. These systems have since been replaced, rewritten, or expanded substantially, but it has always amazed me what a motivated intern can get off the ground in just a few months, without preconceptions about what might be “impossible.” 😌 This year, our summer interns worked directly in complex domains and large existing codebases serving millions of businesses. They’re right in the middle of how the engineering discipline is changing with AI, and there were early and promising signs that it’s now much easier to gain industry-specific context, navigate unfamiliar systems, and build great judgment, all in service of shipping some very ambitious projects. Some examples of what they shipped this summer: 🏦 A new safeguard deactivates closed, frozen, or invalid bank accounts in about five seconds instead of as long as 12 hours, helping prevent avoidable payout failures. 🔑 A new Dashboard tool analyzes how an API key has actually been used and helps a business protect their API keys by replacing its broad access with a narrower set of permissions. 🎨 Checkout studio can preview real products, prices, currencies, and custom branding without overwriting a business’s account-wide styling. The project prepared the experience to expand from a roughly 200-business private preview to more than 140,000 eligible businesses! 🌏 Checkout now flags invalid Australian PayIDs before a PayTo payment is confirmed, so customers can correct mistakes immediately instead of discovering a failed payment later. 🏥 Issuing entered private preview with real-time support for the healthcare-eligible portion of FSA and HSA purchases, allowing issuers to approve the correct amount (including partial approvals). Thanks to all our interns (and their intern mentors!) for a wonderful summer! ♥️☀️
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态Today we shipped the largest revision to MCP since launch. What's in the 2026-07-28 spec release: - Stateless core: The initialize handshake and session IDs are gone. Remote MCP servers can now run behind a plain load balancer. You can deploy on serverless and edge infrastructure, or scale horizontally behind any load balancer. - Extensions are first-class: MCP Apps and Tasks ship as the first two official extensions. - Authorization hardening: We've aligned the spec with real-world OAuth 2.0 and OIDC deployments. While small on paper, this is big for anyone implementing clients or servers. - Updated governance: A formal feature lifecycle, a conformance suite that gates proposals, and SDK tiers. 2026-07-28 and future revisions shouldn't require rewriting transport code. Support for the new spec is rolling out across Claude products. Learn more: https://lnkd.in/ggpGDhdc If you're building with MCP, I want to hear from you. What are you working on?What are you most excited to build? What do you wish the protocol supported? Comment below.👇 Your input helps shape what we build next.MCP 2026-07-28 spec: stateless core, coming to Claude | Claude by AnthropicMCP 2026-07-28 spec: stateless core, coming to Claude | Claude by Anthropic
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态We ship so much it's kinda insane (we only have 3 engineers!) plz send help (you get $5k if we hire them)
-
Kevin Ho 赞了此动态Claude managed agents can now dream to self learn, hit the outcome you specify on its own, and turn into multi-agent fleets.Kevin Ho 赞了此动态Live from Code with Claude: we're launching dreaming in Claude Managed Agents as a research preview. Outcomes, multiagent orchestration, and webhooks are now in public beta. Dreaming is a scheduled process that reviews your agent's past sessions, extracts patterns, and curates memories so they improve over time. Harvey saw ~6x higher task completion rates with dreaming enabled. Outcomes lets you set the bar for quality. You write a rubric, a separate grader checks the output, and the agent iterates until it gets there. Subscribe to webhooks to get notified when it's done. Multiagent orchestration lets a lead agent delegate to specialists that work in parallel on complex jobs. Available today on the Claude Platform. Read more: https://lnkd.in/e4xKcpuR Request access to dreaming: https://lnkd.in/gh7qrp38
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态Last week, GovTech Singapore visited our San Francisco office to spend time with the engineering team at Stripe. We got into how we’re approaching AI engineering, our work on Minions, developer productivity tooling and how we’re thinking about benchmarking AI-powered integrations. Not the usual conversation around payments, money movement, or revenue growth, but a useful reminder that while payments are what we’re known for, much of the differentiation sits behind the scenes in how the engineering teams build and operate. That’s what allows Stripe to move quickly (288 launches at Stripe Sessions!) while still maintaining five nines of reliability. Thanks to Ainsley Escorce-Jones, Kevin Ho, Carol L. for sharing and Pranav Ashok and Andreessen Horowitz for bringing this together. For those curious, you can read more in the miniseries on our engineering blog here: https://lnkd.in/g3v4GS9j https://lnkd.in/gPFx7PDs #AI #Engineering #GovTech #Singapore #Stripe #SEA
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态Just got back from my first Stripe Sessions 2026, where we announced 288 product launches. Thank goodness we now have our public roadmap (stripe.com/roadmap), filterable through Q1 2027 by product and region to keep up. 😄 Had some really great conversations with Stripe users from all over the world, across industries. They came together to share what they’ve built, how they’re approaching growth, monetisation, scale and to learn from each other along the way. Here are a few highlights: • Agentic Commerce: Stripe now enables merchants to sell through AI agents across Google Gemini, OpenAI, Meta, and Microsoft with a single integration. Also introduces a new Machine Payments Protocol for agent-to-agent transactions. • Checkout Studio: A visual workspace to build, monitor, and optimise checkout with AI recommendations. No code changes needed. • Metronome in Stripe: Usage-based billing (powering companies like Anthropic and OpenAI) is now native in the Stripe Dashboard, with real-time credit tracking, alerts, and auto top-ups. • Stripe Treasury: Instant settlements (including weekends), a Stripe card with cashback, and free instant transfers between Stripe accounts. Expanding to 26 markets by year-end. • Stablecoins: Tempo (a payments-focused blockchain), stablecoin-backed card issuing in 100 countries, and global payouts to 150 countries via stablecoin rails. Companies like DoorDash, Klarna, and Ramp are already building on this. • UPI without a local entity: Stripe is currently the only major PSP offering cross-border UPI. One customer saw India revenue increase by 22% in the first month. • Managed Payments: Stripe acts as merchant of record across 195 markets, handling tax, fraud, disputes, and support. If you missed it: Catch the keynote here: stripesessions.com Announcements: https://lnkd.in/gvg85b72
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态We just announced a large raft of improvements at Stripe Sessions. My meta reflections: • It feels that the entire economy is replatforming right now. • Many charts at Stripe are inflecting in quite dramatic ways. What GitHub recently reported for commits we are seeing in economic activity (such as new company formations). • It is increasingly clear that agents will be responsible for most transactions in the not overly distant future. • Stripe was always developer-centric, but AI is making developer-centricity strategic in a new way: agents are even hungrier for good DX than developers themselves are. • Things that we’re launching are increasingly network products at heart. (Instant transfers between Stripe businesses, new kinds of fraud prevention with Stripe Radar, stablecoin payouts to anyone with Link.) "How can we turn Stripe's economies of scale into user benefits?" is increasingly the relevant question. • Between Privy, Bridge, Tempo, and Stripe’s core capabilities, we’re now doing a lot in stablecoins/crypto, and companies like DoorDash, Ramp, Meta, and Klarna are using our crypto stack to deploy meaningful new functionality in production. “But where’s the production use?” is rapidly becoming stale when applied to crypto. • After more than a decade of building, we seem to have hit some kind of critical mass of core platform capabilities such that building new things now feels easier and faster than before. (AI also helps.) We announced Stripe Treasury last year (originally called Financial Accounts); since then, we’ve added multi-currency support, global payouts, card issuance and rewards, and a bunch of other sophisticated functionality. By the end of this year, Treasury will support 15 more currencies and be available to businesses in 160 countries. On the launches themselves, a small selection that I thought were cool, though this is really just a subset: [Cut -- launches instead listed in this tweet: https://lnkd.in/gJNgn7nQ. The LinkedIn post character limit is too low to contain them!] But that’s really just a subset of a subset. (See http:// stripe.com/roadmap for more.) The Stripe team is cooking! And if you’re interested in building the economic infrastructure for this new world, we’re hiring.
-
Kevin Ho 赞了此动态Kevin Ho 赞了此动态Big news: After 4 months of building, we’ve launched new AI-powered insights for platforms, integrated directly with the Stripe network. Platforms can now leverage: 0-100 Fraud Scores: Instant risk assessment for every business and transaction. Explainable AI: Clear insights into why specific accounts are flagged. Unified Views: A complete picture of account activity to mitigate risk at scale. See how it works: https://lnkd.in/gmws5Zmr
工作经历
教育经历
查看Kevin的完整档案
-
浏览共同好友
-
请求引荐
-
直接联系Kevin
其他相似会员
浏览更多动态
-
Daren Guo
Reap • 4335 位关注者
As a Stripe alum, I still read the annual letter with a mix of nostalgia and curiosity. Seeing them process $1.9T in volume (~1.6% of global GDP) really puts into perspective how far payments have come. For anyone interested, it’s worth the read: https://lnkd.in/gtEusJCr Growing up and starting my career in North America, a few things stood out to me about the US market today: Profit concentration is accelerating the largest 50 companies now account for 59% of S&P 500 profits. At the same time, nearly half of US GDP growth in 2025 came from software and data center demand. AI also isn’t just a sector trend but is shaping macro outcomes. And yet, capital access remains uneven. Loans under $1M are still down versus 2010 levels. The biggest players are scaling faster, while smaller operators continue to face structural constraints. To me, that reinforces something I learned at Stripe which is that infrastructure quietly determines who gets to compete. My thoughts are that Stripe helped define the first era of internet payments but the next — programmable, global, increasingly agentic, will be won on execution and trust. Grateful for the foundations I learned as an alum. Excited to be building Reap in this next chapter.
81
-
Justin Callison
4157 位关注者
It is wild how much #agents have changed software development. Excited to see the first of a series of videos land on the https://lnkd.in/gGx9Fzq2 channel, where the team shares some of the things we've learned and some of the things we've built to help Stripe users navigate these changes and drive ever greater success. cc: James Beswick, Manish Kurse, Nilofer Rajpurkar https://lnkd.in/gzejyT2r
27
1 条评论 -
Mathew Varughese
Stripe • 1161 位关注者
Signup forms slow agents down. That's why we built claimable sandboxes: a way to build Stripe integrations without creating an account first. Here's an engineering blog post on how we built them while balancing security, latency, cost, and UX! https://lnkd.in/euJMEdWE
51
3 条评论 -
Steve Kaliski
Stripe • 2989 位关注者
We just rolled out support for WebMCP across Stripe Checkout UIs. In our benchmarking we saw 42% fewer tokens consumed and 38% fewer tools invoked when an agent had access to WebMCP when completing a payment. Our implementation is built on top of all the work we have done (and continue to do!) to optimize checkout experiences -- just now for agents too. In particular, we focused on progressive tool disclosure to only give agents access to the information and actions they need. For example -- the `fill_payment_form` tool only specifies fields that are relevant to the current context (payment method, etc) and `submit_payment` is only revealed when all validations are met. To learn more, read our latest post --> https://lnkd.in/g9kcGxPh
68
6 条评论 -
Jen Lee
Stripe • 3861 位关注者
Stripe has been building toward a world where stablecoins power more and more of the financial products people use every day. Today's a big step in that direction: Phantom is opening the waitlist for Cash Their users will be able to hold a dollar balance, send money instantly, and spend with Phantom or via a card, all from their existing wallet. Behind the scenes, this runs on the Stripe onramp, Bridge stablecoin issuance, virtual accounts, and card issuing. But the key is that the user never sees any of that. They just see Cash they can use. Read more about it here: https://lnkd.in/ebpEebyP
91
2 条评论 -
Kevin Miller
Stripe • 2万 位关注者
Software engineering is changing fast. Supporting the torrid pace of product and feature delivery has required equal (or more!) focus on the right internal dev tools and culture inside Stripe. Recently Payton from Clay spent a day at Stripe's San Francisco headquarters to get a behind-the-scenes look at what this looks like day-to-day. Today's tooling makes it easier than ever for engineers to focus on the strategic, not the mundane: understanding what customers need, building at-scale reliable systems, and presenting it with top-notch quality, craft, and beauty. Worth a watch! https://lnkd.in/gJpDRfr5
67
2 条评论 -
Carol L.
wherever I can yap • 1837 位关注者
AI agents are not going to replace software engineers yet, at least not when building Stripe integrations. The critical gap between our AI reality today and production-ready reality: we sorely lack a stable and realistic validation layer, especially when it comes to financial infrastructure where we require 100% accuracy. So to measure how AI is progressing on that last mile of value, Kevin Ho and I built a set of robust, replicable software engineering environments to benchmark models on their ability to build complete Stripe integrations. They mirror realistic integration scenarios: huge codebases, data migrations, breaking version changes, server and client-side API integrations, and even browser use. The range of results were extremely surprising, and even more exciting is how we now have an experimentation test bed to demonstrably improve how models integrate with Stripe. We’ve open-sourced the full benchmark suite in our AI toolkit: https://lnkd.in/em-aEwMF. Read the blog post: https://lnkd.in/eXAVSHRq.
179
12 条评论