Maker Hyperspace Pirate has created a dive helmet using a 3D printer, tons of fiberglass, and a lot of ingenuity. They transformed an old diving concept into something that works, at least for shallow pool diving. This helmet is modeled like old surface-supplied diving gear. Air rushes in from above and exhaust leaves from below, keeping the diver’s head dry inside a little upside-down plastic bubble.

This project demands precise measurements, so Hyperspace Pirate originally sketched up the dimensions of the helmet on cardboard: 7.5 inches wide, 9.75 inches deep, and 11.25 inches high. Then Onshape software was used to generate a digital model of the object. It was divided into four separate components to accommodate the maker’s 3D printer, a Prusa MK3S, and manufactured with varying layer thicknesses to balance speed and strength. The printer used ABS filament, and after printing, the components were bonded together and encased in fiberglass soaked in epoxy for added strength. He had to be cautious about the sort of resin used because the other type tends to attack the ABS pieces.

Bambu Lab A1 3D Printer, Support Multi-Color 3D Printing, High Speed & Precision, Full-Auto Calibration…

• High-Speed Precision: Experience unparalleled speed and precision with the Bambu Lab A1 3D Printer. With an impressive acceleration of 10,000 mm/s…
• Multi-Color Printing with AMS lite: Unlock your creativity with vibrant and multi-colored 3D prints. The Bambu Lab A1 3D printers make multi-color…
• Full-Auto Calibration: Say goodbye to manual calibration hassles. The A1 3D printer takes care of all the calibration processes automatically…

Three large transparent acrylic panes, one in the front and one on each side, provide visibility and are press-fitted into position and sealed with silicone caulk. The weight of the water at deep forces the windows in and closes them securely. A carry handle on top allows you to move the item about, and there are some useful small fittings for installing lighting or a camera.

Air enters through a small SAE fitting on the top of the helmet, which has a check valve to prevent air from flowing back in if the supply fails. Any extra air leaves at the back through a long, hollow PVC pipe, resulting in constant bubbles. The air supply is handled by a quiet oilless compressor, the MAC100Q type, which produces a consistent 40 liters per minute at up to 2 bar of pressure. This allows us to dive down to approximately 10 meters (33 feet) without running out of air. The rig is powered by a 12-volt, 35 amp-hour lead-acid battery connected to a 2 kilowatt pure sine wave inverter. It’s all mounted on a pool inner tube float, with a flat piece of plywood for stability and a handy diver-down flag.

Ballast weights, which are positioned on either side and held in place by a pair of large steel plates, work to keep the helmet underwater. We began with 30 pounds of weights, but after a few initial tests, we added an additional 10 pounds to achieve neutral buoyancy. A neoprene pad protects the neck against pressure points.

The helmet got a good workout in the pool, with dry runs ensuring proper fit and airflow, and underwater sessions allowing the builder to move around. We’ve had to be careful not to go too deep, partly because the bottom of the pool is a little uneven, but the airflow keeps it nice and cool and dry inside, preventing the windows from fogging or overheating. The compressor ran smoothly, and we were relieved that the helmet did not overheat. There was one peculiarity, however: the exhaust pipe would occasionally give out a small gurgling. It was a little unpleasant at the time, but it turned out to be harmless.

Safety has always been a key priority here, which is why the oilless compressor is a must-have. It means there is no risk of oil vapor or exhaust fumes entering the helmet. And just in case something goes wrong, we have a check valve to prevent the air pressure from decreasing quickly. Despite the fact that the helmet works extremely well for shallow water tasks, Hyperspace Pirate is quick to point out that this is still very much an experimental project, and it is not safe to use in open water or much deeper water without some serious modifications to make it waterproof, as well as a bit of backup air to be safe.
[Source]

US self-driving startup Nuro, which is backed by the likes of NVIDIA, Toyota and Uber, has started testing its autonomous vehicles on Tokyo’s challenging streets, Bloomberg reported. The company, which plans to launch a robotaxi service with Uber and Lucid in San Francisco this year, will be testing a “handful” of vehicles in the city. Human safety drivers will be at the wheel, as is required by Japanese law.

Tokyo presents a challenge for autonomous vehicles, given its narrow, crowded streets and left side of the road driving. “Testing the capability of the autonomy system in such an interesting market with some international complexity really is a good pressure test of what the system is capable of,” said CEO Andrew Chapin. The company’s ultimate goal is to achieve Level 4 autonomy, which allows full self-driving under limited conditions.

Waymo is the other major robotaxi operator testing vehicles in Tokyo in collaboration with Japanese taxi operators Nihon Kotsu and the country’s leading taxi app, Go. It has been operating in the nation since April 2025 in collaboration with Toyota.

Nuro has yet to announce which operators or vehicle manufacturers it will be partnering with, but Chapin said it may not limit itself to autonomous rides. “A universal autonomy platform that can be extended to a lot of different applications and form factors is a bit different than the approach Waymo is taking,” he told Bloomberg. The company previously teamed with 7-Eleven on autonomous deliveries in Mountain View, California.

Uber plans to have up to 100,000 autonomous vehicles including 20,000 robotaxis powered by Lucid and Nuro, with a rollout starting in 2027. It introduced its new vehicle design recently at CES 2026. Uber is also collaborating with Nissan and Wayve with the aim to introduce pilot cars in Tokyo by late 2026.

Update: Story updated with a statement from Stryker confirming they suffered a disruptive cyberattack.

Leading medical technology company Stryker has been hit by a wiper malware attack claimed by Handala, an Iranian-linked and pro-Palestinian hacktivist group.

The medtech giant manufactures a range of products, including surgical and neurotechnology equipment. With over 53,000 employees, Stryker is a Fortune 500 company that reported global sales of $22.6 billion in 2024.

Handala says they stole 50 terabytes of data before wiping tens of thousands of systems and servers across the company’s network, forcing Stryker to shut down in “an unprecedented blow.”

“In this operation, over 200,000 systems, servers, and mobile devices have been wiped and 50 terabytes of critical data have been extracted,” the attackers said. “Stryker’s offices in 79 countries have been forced to shut down.”

This aligns with reports from people claiming to be Stryker employees from the United States, Ireland, Costa Rica, and Australia, who said their managed Windows and mobile devices were remotely wiped in the middle of the night. The attackers have also defaced the company’s Entra login page to display a Handala logo.

A Stryker employee told BleepingComputer the incident began early Wednesday morning, when devices enrolled in the company’s mobile device management system were remotely wiped. The employee said colleagues who had personal phones enrolled for work access also lost data after their devices were reset.

Staff were instructed to remove corporate management and applications from their personal devices, including the Intune Company Portal, Teams, and VPN clients.

Numerous employees also report that the attack disrupted access to internal services and applications, forcing some locations to revert to “pen and paper” workflows after systems became unavailable.

As a result of the attack, Stryker is now working to restore their systems amid a global outage, as first reported by The Wall Street Journal.

“We are experiencing a severe, global disruption impacting all Stryker laptops and systems that connect to our network,” Stryker told employees in Cork, Ireland, according to local media.

“At this time, the root cause has not yet been identified. We are actively engaged with Microsoft and treating this a critical, enterprise-wide incident,” the company added in a message sent to employees in Asia.

Handala (also known as Handala Hack Team, Hatef, Hamsa) first surfaced in December 2023 as a hacktivist operation linked to Iran’s Ministry of Intelligence and Security (MOIS) that targets Israeli organizations with destructive malware designed to wipe Windows and Linux devices.

They are also known for stealing sensitive data from victims’ compromised systems and publishing it on the group’s data leak portals.

Stryker confirms cyberattack

Malware is getting smarter. The Red Report 2026 reveals how new threats use math to detect sandboxes and hide in plain sight.

Download our analysis of 1.1 million malicious samples to uncover the top 10 techniques and see if your security stack is blinded.

Advertisement

Download The Report

from the oh-buc-off dept

Alright, I think it might be time for a wellness check on the people running Buc-ee’s.

I realize that these chain of gas and convenience stores has a strange cult following in the south. I won’t pretend to understand why that is, but whatever. Unfortunately, the company also appears to be run by a bunch of trademark bullying jackwagons. I’ve referred to Buc-ee’s as the Monster Energy of gas stations, because the company appears to think that trademark law allows it to own the concept of a cartoon animal mascot in any tangential industry. They have bullied and/or sued many, many companies under this premise. Because most of its victims are smaller companies, they have gotten a lot of settlements out of these bullying efforts.

But those settlements don’t make the bullying legitimate. Buc-ee’s views on what trademark law allows it to own and control are fantasy. They’re still out here doing their bullying thing, though, with the latest example being its decision to sue a company that runs a gas station called “Mickey’s”. I’ve embedded the suit below, but here is a sample of the claims in the filing made against the gas station chain.

Like the Buc-ee’s Marks, Defendant’s Logos incorporate a cartoon animal facing right with wide eyes and a smile, overlaying a round background…also uses red as a predominant color in its interior and exterior signage, as well as employee uniforms and anthropomorphic representations of its cartoon moose mascot…also uses red as a predominant color in its interior and exterior signage, as well as employee uniforms and anthropomorphic representations of its cartoon moose mascot.

Consumers are likely to perceive a connection or association as to the source, sponsorship, or affiliation of the parties’ products and services, when in fact none exists, given the similarity of the parties’ logos, trade channels, and consumer bases.

Advertisement

And here, dear readers, is the very similar branding that the lawsuit references.

Once again, as with past Buc-ee’s trademark suits, the claims simply fall apart on inspection of the evidence. These logos are not similar. They don’t use the same overall color schemes. They feature easily distinguishable cartoon animals as mascot. A beaver is not a moose, which is a sentence I never thought I’ve have to type out on a keyboard. Likewise, a hexagon is not round, another thing I’d never thought I’d have to write. This is all very, very stupid, and not at all concerning from a customer confusion standpoint.

Despite that, the suit alleges that Mickey’s has “used” the Buc-ee’s logos to enrich themselves. It’s bonkers. In addition, Buc-ee’s has petitioned the USPTO to cancel the trademark registrations Mickey’s has for its branding.

Why is this company so beloved? They truly seem like craven bullies above all else. None of this is trademark infringement and I certainly hope the owners of Mickey’s are prepared to fight this fight. Because Buc-ee’s doesn’t somehow have a monopoly on cartoon character mascots. Not for its industry, never mind others.

Filed Under: beaver, moose, trademark

Companies: buc-ee’s, mickey’s

This article is crossposted from IEEE Spectrum’s careers newsletter. Sign up now to get insider tips, expert advice, and practical strategies, written in partnership with tech career development company Parsity and delivered to your inbox for free!

How to Keep Your Engineering Skills Sharp in an AI World

Engineers today are caught in a strange new reality. We’re expected to move faster than ever using AI tools for coding, analysis, documentation, and design. At the same time, there’s a growing worry in the background: If the AI is doing the work, what happens to my skills?

That concern isn’t just philosophical. Research from Anthropic, the company behind Claude, has suggested that heavy AI assistance can interfere with human learning—especially for more junior software engineers. When a tool fills in the gaps too quickly, you may deliver working output without ever building a strong mental model of what’s happening underneath.

More experienced engineers often feel a different version of this anxiety: a fear that they might slowly lose the hard-earned intuition that made them effective in the first place.

In some ways, this isn’t new. We’ve always borrowed solutions from textbooks, colleagues, forums, and code snippets from strangers on the internet. The difference now is speed and scale. AI can generate pages of plausible solutions in seconds. It’s never been easier to produce work you don’t fully understand.

I recently felt this firsthand when I joined a new team and had to work in a codebase and language I’d never used before. With AI tools, I was able to become productive almost immediately. I could describe a small change I wanted, get back something that matched the existing patterns, and ship improvements within days. That kind of ramp-up speed is incredible and, increasingly, expected.

But I also noticed how easy it would have been to stop at “it works.”

Instead, I made a conscious decision to use AI not just to generate solutions, but to deepen my understanding. After getting a working change, I’d ask the AI to walk me through the code step by step. Why was this pattern used? What would break if I removed this abstraction? Is this idiomatic for this language, or just one possible approach?

The shift from generation to interrogation made a massive difference.

One of the most powerful techniques I used was explaining things back in my own words. I’d summarize how I thought a part of the system worked or how this language handled certain concepts, then ask the AI to point out gaps or mistakes. That process forced me to form my own mental models rather than just recognizing patterns. Over time, I started to build intuition for the language’s quirks, common pitfalls, and design style. This kind of understanding helps you debug and design, not just copy and paste.

This is the core mindset shift engineers need in the AI era: Use AI to accelerate learning, not to replace thinking.

The worst way to use these tools is also the easiest: prompt, accept, ship, repeat. That path leads to shallow knowledge and growing dependence. The better path is slightly slower but more durable. Let AI help you move quickly, but always come back and ask, Do I understand what I just built? If not, use the same tool to help you understand it.

AI can absolutely make us faster. Used well, it can also make us better at our jobs. The engineers who stay sharp won’t be the ones who avoid AI, they’ll be the ones who turn it into a collaborator in their own learning.

—Brian

When war strikes, critical power infrastructure is often hit. Engineers in Ukraine have risked their lives to keep electricity flowing, and some have been hurt or killed in the dangerous wartime conditions. One such engineer, Oleksiy Brecht, died on the job in January. “Brecht’s life and death are a window into the realities of thousands of Ukrainian engineers who face conditions beyond what most engineers could imagine,” writes IEEE Spectrum contributing editor Peter Fairley.

Read more here.

The semiconductor industry needs more engineers to build the chips that power our daily lives. To help expand the talent pool, the industry is testing new approaches, including training software engineers to design hardware with the help of AI tools. All engineers will still need to have an understanding of the fundamentals—but could computer science students soon apply their coding skills to help design hardware?

Read more here.

Effective writing and communication are among the most important skills for engineers looking to advance their careers. Though often labeled a “soft skill,” clear communication is essential in both academia and industry. IEEE is now offering a course covering key writing skills, ethical use of generative AI, publishing strategies, and more.

Read more here.

TL;DR

Moz Pro now includes AI-powered keyword suggestions, an AI Visibility feature in open beta, and the industry-standard Domain Authority metric, all starting at $39/month on annual billing. A 7-day free trial is available on Standard and Medium plans, and a free Moz Community account gives permanent access to limited versions of Keyword Explorer and Link Explorer at no cost. In a market where most SEO platforms charge $100+/month before you see any data, Moz Pro is the most affordable way to bring AI into your SEO workflow.

Disclosure: This article contains affiliate links. If you click through and make a purchase, we may earn a commission at no additional cost to you. This helps support our content. All opinions expressed are our own.

Something fundamental changed in SEO this year, and most businesses haven’t caught up yet. Google’s AI Mode is now generating answers from sources that don’t even appear in the traditional top 10 results. Moz’s own research across nearly 40,000 queries found that 88% of AI Mode citations come from URLs that aren’t ranking in the organic SERP for the same query. The old playbook of ranking for a head term and collecting the traffic is no longer the whole game. Topic authority, content structure, off-site visibility, and brand presence across the web now determine whether your business shows up when AI assembles its answers.

For anyone trying to navigate this shift, the tooling question gets uncomfortable fast. The major SEO platforms run between $100 and $250 a month, and most of them lock their newest AI features behind mid-tier plans that cost even more. Ahrefs Lite starts at $99/month. Semrush Pro starts at $139.95/month. If you’re a freelancer, a small team, or a business that knows it needs professional SEO data but can’t justify enterprise pricing, those numbers create a gap between free tools that lack depth and paid platforms that assume a corporate budget.

Moz Pro fills that gap, and in 2026, it’s doing it with AI. Starting at $39 per month on annual billing, it’s the lowest entry price among the three major platforms, and it now includes AI-powered keyword suggestions on every plan, an AI Visibility feature currently in open beta, and the same Domain Authority metric that the entire marketing industry already speaks fluently. Start your free trial at moz.com

What Moz’s AI features actually do (and why they matter right now)

Moz has been in the SEO business since 2004, longer than Ahrefs (2011) and Semrush (2008). That longevity produced Domain Authority, the 1-100 score predicting how likely a domain is to rank in Google results. DA is used in agency reports, PR tools like Cision, journalist outreach platforms, and client pitches across the industry. When someone says “our DA is 52,” everyone in marketing knows exactly what that means. And Moz Pro is the only platform that calculates it natively.

But the more interesting story in 2026 is what Moz is building on top of that foundation. Keyword Explorer, the platform’s centerpiece tool, now includes AI-powered keyword suggestions that go beyond simple stem matching. Instead of generating variations of the word you typed in, the AI pulls from semantic relationships and actual search behavior patterns to surface keyword ideas you wouldn’t find through traditional methods. It also includes a Priority score that combines search volume, keyword difficulty, and organic click-through rate into a single number, so instead of toggling between three data columns and doing mental math, you get a clear recommendation of which keywords are worth targeting right now.

Then there’s the AI Visibility feature, currently in open beta. This is Moz’s response to the measurement problem that the entire industry is grappling with: if AI-generated answers are pulling citations from sources beyond the traditional top 10, how do you know whether your brand is being mentioned? The setup is simple. Enter your brand name, add related terms, input competitors, and Moz AI generates prompts based on queries users are actually asking about your brand. You get a dashboard showing citation frequency, competitive positioning, and visibility trends across generative search results.

Brand Authority rounds out the AI-adjacent features. It’s a metric that scores your brand’s overall search presence, not just backlinks, but how often your brand appears in search results relative to competitors. For businesses that care about visibility beyond traditional keyword rankings, this adds a dimension that rank tracking alone can’t capture.

These aren’t cosmetic additions. When Google’s AI Mode is assembling answers from YouTube, Reddit, LinkedIn, and niche authority sites (and Moz’s own research confirms that it is), the tools that help you understand and optimize for that broader landscape are the ones worth paying for.

The full toolkit and what it costs

AI features aside, Moz Pro earns its subscription through how it connects its tools into a workflow that makes sense for people managing SEO alongside other marketing responsibilities. Site Crawl runs automated weekly audits, flagging broken links, duplicate content, missing meta tags, and crawlability issues, all severity-ranked so you fix the high-impact problems first instead of drowning in a list of 400 warnings. Link Explorer provides backlink analysis using an index of approximately 45 trillion links, and every link gets a Spam Score to help you identify toxic backlinks. The Link Intersect tool shows which domains link to your competitors but not to you, giving you a ready-made prospecting list. Rank tracking updates weekly with mobile and desktop segmentation and local search tracking.

Moz Pro keeps its pricing transparent across four plans, all with a 20% discount on annual billing. The Starter plan at $49/month ($39/month annually) covers 1 campaign, 50 tracked keywords, and 20,000 pages crawled. For a single website that needs professional keyword research and basic rank tracking, this is the most affordable entry point in the all-in-one SEO market. The Standard plan at $99/month ($79/month annually) jumps to 3 campaigns, 300 tracked keywords, and 400,000 pages crawled. This is where most users land, and it’s also one of two plans eligible for the 7-day free trial. The Medium plan at $179/month ($143/month annually) scales to 10 campaigns and 1,500 tracked keywords, and is the second plan eligible for the free trial. The Large plan at $299/month ($239/month annually) maxes out at 25 campaigns and 3,000 tracked keywords for agencies and large in-house teams.

For comparison, Ahrefs Lite starts at $99/month with no free trial at all. Semrush Pro starts at $139.95/month with a 7-day trial on that single plan. Moz undercuts both on price and offers a more flexible entry path. For a full breakdown of how these platforms compare, Tekpon’s Moz Pro review and Moz Pro pricing comparison cover the details.

How to get the best deal right now

Moz gives you two free entry points, which is more than most competitors offer. The 7-day free trial provides full access to either the Standard or Medium plan. Credit card required to start, but you can cancel at any time during the trial and keep access for the remaining days. Seven days is enough to run a full site audit, pull keyword research for a content plan, test the AI-powered keyword suggestions, check your AI Visibility dashboard, analyze your backlink profile, and set up rank tracking. You can build an entire SEO action plan before paying a dollar.

The free Moz Community account is permanent, no trial period, no credit card. It gives limited access to MozBar (the Chrome extension that overlays DA and PA on any webpage), Keyword Explorer (10 queries per month), and Link Explorer (10 queries per month). For someone who only needs occasional DA checks or quick keyword lookups, this may be all you need.

The optimal strategy is to start the 7-day free trial on the Standard plan ($79/month annual). Test the AI features, run your audits, pull your keyword research. If the data proves useful (and for most sites doing any meaningful SEO work, it will), lock in the annual rate for a 20% savings. Standard drops to $79/month, saving $240 over the year. Medium drops to $143/month, saving $432. No promo codes needed. Registered 501(c)(3) nonprofits can access a dedicated discount page directly from Moz. See all plans at moz.com

Moz Pro Standard at $79/month (annual) includes 3 campaigns, 300 tracked keywords, AI-powered keyword suggestions, AI Visibility in beta, and a full site audit engine, at less than half the price of comparable Ahrefs and Semrush plans. Start your free trial at moz.com

The search landscape in 2026 rewards brands that show up across the entire ecosystem, not just in the ten blue links. Moz Pro is one of the few platforms that’s building tools for that reality while keeping the price accessible enough that you don’t need an enterprise budget to use them. The AI features are already live. The free trial is already available. The only cost of waiting is the visibility you’re not tracking yet.

For decades, software companies designed their products for a single type of customer: a human being staring at a screen. Every button, menu, and dashboard existed to translate a person’s intention into a machine’s action. But a small startup based in San Francisco and Zurich believes that era is ending — and that the future belongs to companies that build software not for people, but for the artificial intelligence agents that increasingly act on their behalf.

Manufact, a three-person company that emerged from Y Combinator’s Summer 2025 batch, announced in February that it raised $6.3 million in seed funding led by Peak XV, the venture capital firm formerly known as Sequoia Capital India and Southeast Asia, which now manages more than $10 billion in assets. Liquid 2 Ventures, Ritual Capital, Pioneer Fund, and Y Combinator also participated in the round, alongside angel investors including the co-founder and chief operating officer of Supabase.

The company’s thesis is deceptively simple and potentially enormous: as AI agents take over more of the work that humans perform inside software applications — filing expense reports, managing customer support tickets, writing code, booking travel — every software product on earth will need a new kind of interface designed specifically for those agents. Manufact is building the open-source tools and cloud infrastructure to make that transition possible.

“Software products are already being accessed by and will be accessed mainly by AI agents, or by users through chat interfaces,” Luigi Pederzani, co-founder and co-CEO of Manufact, said in an interview with VentureBeat. “That’s our bet. That’s our thesis. And that’s what we are really rooting our company on.”

How Anthropic’s Model Context Protocol became the universal standard for AI agents

To understand Manufact, you first have to understand the technology it is built on: the Model Context Protocol, or MCP, an open standard introduced by Anthropic in late 2024 that has rapidly become the dominant way for AI agents to communicate with external software tools and data sources.

Before MCP, connecting an AI agent to a company’s software required custom integration work for every single tool — a bespoke connector for Slack, another for Salesforce, another for a database. It was tedious, expensive, and fragile. MCP standardized this process into a single protocol, functioning as what CIO magazine recently called “the USB-C of AI” — a universal connector that lets any AI model plug into any software system through a single, consistent interface.

The adoption has been explosive. In December 2025, Anthropic donated MCP to the Linux Foundation’s new Agentic AI Foundation, co-founded with Block and OpenAI, with support from Google, Microsoft, Amazon Web Services, and Cloudflare. More than 10,000 active public MCP servers now operate across the ecosystem. ChatGPT, Cursor, Google Gemini, Microsoft Copilot, and Visual Studio Code all support the protocol. Enterprise-grade deployment infrastructure exists from AWS, Cloudflare, Google Cloud, and Microsoft Azure. An estimated 7 million downloads of MCP servers occur every month.

“Great protocols are as good as their adoption,” Pederzani said, drawing a comparison to the mobile revolution. “We saw the same transition with mobile, right? In the beginning, companies were just creating a pretty simple mobile app. Who would have bought a hotel or a flight or used a bank account from a mobile app? But as time passed, the web became mobile first. What we think is that software products will be MCP first, or chat first.”

The stakes are high. The global AI agents market reached $7.84 billion in 2025 and is projected to surge to $52.62 billion by 2030, according to industry analysts. The MCP Dev Summit, the largest conference dedicated to the protocol, takes place April 2–3 in New York City under the Linux Foundation’s banner, with speakers from Docker, Workato, and major cloud providers — and Manufact will be among the companies presenting.

Two Italian founders, a Zurich co-working space, and an open-source library that went viral

Manufact’s origin story reads like a case study in the power of open-source communities to validate a startup idea before a single dollar of venture capital is raised.

Pietro Zullo and Luigi Pederzani, both originally from Italy, met at a co-working space in Zurich — the same space that produced Browser Use, Bloom, and other startups that went through YC in previous batches. Zullo was studying at ETH Zurich; Pederzani was working at Morgen, an ETH spin-off AI startup used by teams at Spotify, GitHub, and Linear, after leading a 12-engineer team at Accenture Switzerland. Both were winding down previous projects in early 2025 when MCP launched.

“We both wrote agents in the past, and it was such a mess to write the tools, the integrations,” Zullo recalled. “When MCP came out, it looked like the perfect fit for what we were trying to do. But only Cursor, Claude Code, a few closed-source applications allowed you to actually use the protocol. I don’t think I’m going to do groceries or browse the internet or check my emails from Cursor — it’s like, not the right code, right? So we wrote an open-source library to basically do what you could do in Cursor with MCP servers, but on your own machine, on your own application, in your own terms.”

They called the library mcp-use, with a slogan that resonated across the developer community: “Connect any MCP to any LLM in six lines of code.” The repository attracted 2,000 to 2,500 GitHub stars within weeks. Today, the SDK has surpassed 5 million downloads and 9,000 GitHub stars. Organizations including NASA, Nvidia, and SAP use the library, and Manufact claims that 20 percent of the US 500 have experimented with it.

“The amount of power that you can put in six lines of code was really staggering,” Zullo said. The pair applied to Y Combinator on the day of the deadline. “We were super spontaneous because we had this open-source vibe and just enjoyed the process. We had so much energy from the community that was lifting us up, and we knew it was going to be fine.”

Inside Manufact’s plan to become the ‘Vercel for MCP’ — from SDK to cloud in 60 seconds

Manufact’s strategy borrows directly from the playbook that turned Vercel into a multi-billion-dollar company by providing hosting and developer tools for front-end web applications. The analogy is deliberate: just as Vercel made it trivially easy to deploy a Next.js app, Manufact wants to make it trivially easy to build, test, and deploy the MCP servers and MCP apps that AI agents need to interact with software.

The company offers three core products. First, the open-source mcp-use SDK, available in both Python and TypeScript, lets developers spin up a fully functional AI agent connected to MCP tools in as few as six lines of code. It supports any large language model, including local models, and has integrations with LangChain and other popular frameworks. Second, a built-in inspector and testing suite allows developers to visually debug their MCP servers in a browser, view raw JSON-RPC traffic, and test tool execution in a sandbox — without connecting to a live AI agent. Third, the Manufact Cloud platform handles deployment, scaling, authentication, access control, and observability, allowing teams to go from a GitHub push to a production MCP server in under 60 seconds.

“As software becomes more agentic, the hard part isn’t the model anymore — it’s everything around it,” Zullo said. “We started Manufact because developers were spending too much time on plumbing instead of building and shipping their products.”

The company has also moved aggressively into MCP apps, a newer extension of the protocol that allows developers to render interactive user interface components — React widgets, data visualizations, input forms — directly inside chat clients like ChatGPT and Claude. Manufact’s SDK lets a developer scaffold an MCP app with a single terminal command, edit React widgets, and deploy to ChatGPT in under a minute. This positions the company at the center of a potentially massive new distribution channel: ChatGPT alone has more than 800 million users.

5 million downloads, zero revenue, and a crowded field of cloud giants

Every open-source company faces the same fundamental tension: the community that makes the project valuable is not the same thing as a paying customer base. Manufact has been candid about this challenge.

Pederzani said the company made a deliberate decision after Y Combinator to focus entirely on the open-source product and community, rather than rushing to monetize. “A lot of open-source projects jump immediately on the monetization part and kind of betray the community,” he said. While NASA, Nvidia, and other prominent organizations use the SDK, Pederzani acknowledged they are not paying customers. Manufact’s target is to reach $2 million to $3 million in annual recurring revenue by the end of 2026, which would position it for a Series A fundraise.

The competitive landscape is crowding fast. AWS, Cloudflare, Vercel, and Docker have all launched MCP hosting features. But Manufact’s founders argue they sit in a complementary position relative to the model providers. “Anthropic and OpenAI are betting that their own chat products — Claude and ChatGPT — will become the primary interfaces through which people access all software,” Pederzani said. “If that bet plays out, we will serve these systems. That’s going to be massive.”

Why software companies without MCP servers risk becoming “dumb databases” for AI agents

Behind Manufact’s optimism lies a darker observation about the software industry that gives their pitch urgency. Pederzani argued that companies that fail to make their products accessible to AI agents risk being reduced to “systems of record” — dumb databases that agents query but that no longer own the user experience or the customer relationship.

“Now we have customers that come to us and say that their customers are choosing to adopt their product over a competitor because they offer an MCP server,” Pederzani said. “At the same time, there is a threat here that could put companies to become just systems of records. And this is really something that a lot of companies are scared of.”

In late February, Manufact co-hosted what it called the largest MCP apps hackathon to date at Y Combinator’s headquarters in San Francisco. The event drew 650 applications and 300 builders. OpenAI, Cloudflare, and Anthropic all sponsored it. Perhaps the most telling detail: eight employees from Anthropic attended — more people than Manufact’s own three-person team. The model providers, it appears, view Manufact as an ally rather than a threat.

Three employees, $6.3 million, and the ambition to capture a share of every AI tool call on Earth

For all its momentum, Manufact faces significant headwinds. The company has just three employees and has not yet demonstrated a scalable revenue model. Its most high-profile users are not paying customers. The $6.3 million seed round provides limited runway in an industry where infrastructure companies often require substantial capital to reach profitability. And the cloud providers that have launched MCP hosting features already own the customer relationships and billing infrastructure that enterprise buyers rely on.

But when asked what success looks like in two years, both founders pointed to a single metric: the percentage of global AI tool calls that flow through their infrastructure. “Our metric is the global tool calls or servers that run on Manufact — how many tool calls are passing through Manufact, made by agents,” Pederzani said. “Like Stripe is doing for the global GDP. We’re going to win if we can get a great number for it.”

The Stripe analogy is ambitious — Stripe processes hundreds of billions of dollars annually and is valued at roughly $90 billion — but it captures the scope of what Manufact’s founders believe is at stake. If MCP becomes the universal standard through which AI agents interact with all software, the company that provides the infrastructure for building and deploying MCP servers could occupy a position of outsized influence.

“In the end, what matters is to make something agents want,” Zullo said, riffing on Y Combinator’s famous dictum to “make something people want.” “What we’re focusing on and what we’re building is to help this transition of building for agents instead of building for humans.”