Forge

The Teaching Hospital for SoC

Our students build, and they build well. They ship working software in their courses, in Orbital, in hackathons, in open-source projects that people the world over depend on. The talent is here, and so is the appetite. What a field this strong is ready for next is the one thing medicine has long taken for granted: a place to form its builders on real work, end to end, the way a teaching hospital forms a young doctor on real cases, beside someone who has done it before.

That place is what I am proposing, and I believe we are the school to build it. A center we run, open to every one of our students, where they earn their way in and up by building. A student comes in not on a promise but on a piece of working software and a real contribution to something people already use. The work speaks for her.

There was a time when the right way to back a young builder was to hear her out and grant her the means, because building was expensive and a good plan was the best evidence anyone had before the fact. That was sound, and it gave us a great deal. What has changed is the cost: a student today can build the thing itself in the time it once took to describe it. So we can ask to see the work, not the promise of it, and that is the fairer test, and the more generous one. The door is open to every student, and the bar at the door is real. Open at the door, high at the bar.

Inside, our students sit down with a real business, come to understand what it truly needs, and build it, with whatever technology fits, especially but not only AI. The technology is never the point. The point is the builder we form, and the work that forms her is real.

And technology is now the substrate under every industry, more than it has ever been. So a place that forms builders on real work does not only serve the builders; through them it serves organizations across every sector, because the firms that come to us for those builders come from every walk of the economy. Forming the builder well is, in the same motion, how the whole economy gets served.

Formed on Real Work, Answerable for the Outcome

A Forge engagement begins the way a teaching-hospital case does, with something real, and someone counting on how it turns out. An SME comes to us with a genuine need: it knows its business and its problem, and it does not know the technology, which is exactly why it has come. From there the work runs the whole way. The student understands the problem, designs the response, builds it and puts it in place, and then stays with it, running and maintaining it over time, because we stand behind what we ship. And somewhere in the middle of that arc the thing happens that tells you a real engagement is under way: he surfaces a need the client never thought to name, because he is now close enough to the work to see what the client could not.

Around that work stand a practitioner and one of our own faculty, together, and what they owe runs in two directions at once. They protect the SME, whose business is riding on the result. And they form the student, who came here to become a particular kind of builder. This is the heart of the teaching-hospital idea, and the reason the word for it is mentorship, not inspection. The senior doctor on the ward is not there to check a junior's work for a customer, but to make a doctor of him, and the patient is kept safe in the very same motion. Both duties are honored at once, and neither is traded away for the other.

And there is something here for our faculty that is theirs alone. The problems a student meets inside a living business are rarely the tidy ones, and the ones worth solving become real material for a faculty member's own research, a line of inquiry that arrives with a working context, and a real person already waiting on the answer.

Deeper Than Any Internship Can Reach

An internship can teach a great deal, and the best of them do. A student joins a defined team, takes on a defined task for a season, and learns how one company works from the inside. Forge is not a substitute for that, and it is not meant to be. It reaches something a placement was never built to reach.

Think of where so many of our most ambitious students long to go: an early-stage startup. It is a formative place to be, and the pull toward it is the right instinct. But an early-stage company is, by its nature, a singular thing, one team improvising its way through one particular problem at one particular moment. She will learn an enormous amount there, and much of what she learns is bound to that company and that moment, and does not carry cleanly to the next thing she does. That is the nature of the stage, not a failing of the people who run it.

Forge gives her the other thing. She owns the problem from end to end, not a slice of it. The stakes are real, because a real business is waiting on the result. The mentorship stays with her, term after term, instead of ending with the season. And across many engagements she comes to know many organizations rather than one, so what truly generalizes about how a business works, and how technology takes hold inside it, begins to separate from the accidents of any single place.

Because Forge runs alongside our academic calendar, she goes further every term. She starts out building under a close eye. She becomes someone whose word is trusted, as a client adopts what she made and a maintainer merges what she wrote. And as the harder engagements ask her to lead, she begins to help run Forge itself, finding the businesses we serve and learning, first-hand, what it takes to run one. Each stage is earned by the body of real work behind it, and the proof is there for anyone to see. She never has to wait to be told she is ready; by the time she leads, everyone already knows she is.

The Hands That Complete the National Effort

Singapore has put real weight behind AI, and you can see it. There is a national plan worth more than a billion Singapore dollars. AI Singapore has built a generation of training and enablement into genuine capability across the workforce. And OpenAI has opened its first Applied AI Lab outside the United States right here, alongside a bootcamp for forward-deployed engineers, around 20 May 2026. Awareness, funding, ready-made tools, training: the ground has been well prepared, and the firms best placed to use it are already moving.

For the largest firms, that is enough; they have the engineers to turn a capable tool into a working system inside the business. For our smaller firms the ground is prepared but the building has not yet begun, and the figures show just where the line falls. In 2024, AI adoption among SMEs ran at 14.5 percent, against 62.5 percent for the larger firms.

The support that reaches our smaller firms is real, and it arrives in four forms: grants that make AI cheaper, catalogs of pre-approved tools, training and enablement, which is where AI Singapore does so much of its work, and advisory services. Every one of them helps. And not one of them builds the thing itself. A grant does not write the integration; a catalog points a firm toward a ready-made tool rather than the custom one it needs; a course teaches a skill but does not build the system; advice produces a recommendation, not working software.

National program	Who it reaches	Form of support
For SMEs
SMEs Go Digital · GenAI Navigator	SMEs	Grants (up to 50%) + pre-approved tools
CTO-as-a-Service · GenAI Sandbox	SMEs	Advisory + trial environments
AI Singapore · AI for Industry, courses	SMEs / workforce	Training + enablement
Productivity Solutions Grant	SMEs	Subsidy for pre-scoped solutions
For large firms
Enterprise Compute Initiative (S$150m) · GenAI Playbook	Large / mature firms	Compute + bespoke partnerships
Champions of AI	Selected leading firms	Capability support

There is a deeper reason behind all of this, and it is worth naming. A generation ago, almost every firm of any size kept someone technical on staff: the person who stood up the email server, wired the office network, and kept the systems running. The cloud quietly retired that role. Email, storage, and software are subscriptions now, bought rather than built, so the in-house technical person gradually disappeared from the small business and was never replaced. Training assumes someone inside who can take a new skill and build with it, and in most small firms that someone is no longer there. What the business needs is not another course; it is a builder, someone to do the work the in-house engineer once did.

What stands between those two numbers is not awareness, and it is not access to tools. It is people who can build. A capable tool only becomes a working system when someone who understands that particular firm, its data and its habits, sits down and builds the fit. That is the piece the whole effort is still waiting on: implementation, a pair of hands that builds, and it is exactly the builder we form. AI Singapore has built the capability; what it leaves room for is the person who turns that capability into something that runs. And we are the school to produce them, a place already full of people who build well, and so the natural one in Singapore to form the builders our firms need.

Forge is broad-based by design, and it begins with SMEs because they are the one segment it can serve without building machinery first. From there it expands deliberately outward, earning the apparatus each new segment demands. Larger firms need systems bolted on before we can serve them well, the standing to bid for their projects and security certifications like SOC 2, none of it trivial to settle up front. The closed and regulated sectors, healthcare and defense among them, feel the same talent shortage but for a different reason. The SME gap is a gap of scale: the talent exists, but a firm too small to keep a builder on staff cannot afford or attract one. The closed-sector gap is a gap of access: these sectors are often large and well-funded, yet builders are kept out, defense by clearance and secrecy, where classified work can never be shown and so can never earn the public reputation that draws the best builders, and by slow procurement; healthcare by domain depth, patient-data and clinical-safety regulation, and an institutional caution that is entirely warranted. An SME cannot afford to bring a builder in; a hospital or a defense agency cannot easily let one in.

Forge is suited to the hard sectors too, and for reasons that follow straight from the model. Supervision is precisely how you place a formed-but-junior builder safely into a high-stakes, regulated setting, which is the whole point of a teaching hospital; a university is a counterparty these sectors can trust on governance, data handling, and security where a freelancer or a young startup is not; and for healthcare the teaching-hospital parallel becomes literal. I want to be honest that this is eventual, and it needs adaptation, not day one. Defense work cannot sit inside Forge's open, shared-publicly model, so a defense track would be a differently governed wing rather than part of the open cohort, and healthcare needs its regulatory and data scaffolding in place first. So we start where the model runs cleanest, the open SME work, earn the record there, and extend from strength. That is disciplined judgment about sequence, not a ceiling on reach.

The Most Wanted Job in Technology

The builder that Forge forms is rarer than a strong engineer, and the years ahead are going to ask for a great many of him. He sits with an organization that knows its problem and not the technology. He finds the specification it never thought to ask for. He builds it. And he stays with it until it is genuinely in use, because the center that ships a thing is the same one that keeps it running.

The industry already has a name for the closest version of this person: the forward-deployed engineer, who works inside the customer's world rather than at arm's length from it. Right now it is the hottest job in technology. That is the answer to why now: the work we have our students doing is the very work the market is most hungry for. And I want to be honest about the size of the claim. It is not that he leads every change he touches; it is something narrower, and more durable, than that. He understands the organization, he builds the fit, and he sees the work through. That is the rare and wanted thing.

So every student who comes through Forge, C-Innovator or not, gains from it, and walks out with a real edge over computing graduates from other schools as he steps into that market. The demand is already here. Forge is how our students meet it ready.

Real Problems, Made Into Public Knowledge

Every engagement starts as a real question no one has answered yet: what does this particular firm need, and what will it actually take to make it work where they are. Answer that question enough times, across enough firms, and the patterns begin to show, what holds up in practice and what quietly does not. Those patterns, the ones that travel, become a body of knowledge we can share openly, with each client's own code and specifics kept its own. The firm's build stays the firm's; the general lesson becomes everyone's.

It is, in miniature, the same work a frontier applied lab does: take on real deployments, learn what generalizes, and share what holds. A lab like OpenAI's does exactly this, and does it superbly. What sets us apart is not the work but its economics. A frontier lab's economics carry it toward the largest contracts; the small firm with a stubborn, particular problem rarely clears that bar. We are built to serve exactly those firms, because our students power the work, and that is what lets us be the one applied lab able to reach the long tail of small and medium businesses, the bulk of the firms in any economy.

I mean applied research in the sense a lab like OpenAI means it: its Applied AI Lab and forward-deployed model, where the research comes from doing real deployment work, not in the traditional university sense of an investigator-led, grant-funded, publication-first agenda. That distinction matters, because it is exactly what keeps Forge from treading on our own School's researchers and its research centers. What they do is deep, funded, IP-oriented research, chosen for its novelty and pursued with a few large partners, and it is some of the best work we do. What Forge produces is different in kind: the broad, public, practice-based field knowledge that falls out of doing real implementation across many SMEs at volume, what actually breaks when technology goes into a small firm, which formal research systematically under-samples because SMEs are not fundable research partners.

So Forge does not rival the centers; it feeds them. It sits upstream of them, a live stream of real problems, real deployment data, and standing relationships with many organizations of the kind the centers are usually starved of, a subset of which can graduate into their formal partnerships. The currency is different, public field knowledge and student capability on one side, papers and IP on the other, so the outputs do not compete. Everyone comes out larger.

And because we are a university center rather than any one company's lab, we can be neutral ground for every frontier lab at once, reaching the long tail none of them will serve directly. That is a genuine win on every side: the labs see their tools land in a market they cannot reach on their own, the businesses get served, and our students learn on problems no one has solved yet. The claim is a modest one, and worth keeping modest. Not a research institute. A growing, practical, public record of what works, de-identified and shared, built one real engagement at a time.

It Pays Its Own Way

The model is built so the center carries itself. A dedicated academic lead of practice-professor standing runs Forge as the work itself, not as one more duty stacked onto a full teaching load. The engagements are paid at the rate this work already commands in the market, three thousand to five thousand Singapore dollars a month, so the center meets its student stipends and its lean running costs out of the value it creates. At steady state, it asks our school for no recurring budget. Through the first cohort it will need some startup help on the administrative side, and I would rather name that plainly now than have it surface later.

The usual way a center like this gets started is with a corporate or government sponsor, and that route stays open to Forge; it would add standing and reach, and I would welcome it. But Forge is built not to depend on it. It pays its own way from the work it does, and sponsorship, if it comes, is a boost rather than the foundation. That is deliberate, and it protects the neutral ground: because the center funds itself at the center level and unrestricted, no sponsor is steering which firms we serve or which projects we take.

An educator will want to ask, all in one breath, who owns the software, what happens when a build fails, whether our students are protected rather than worked cheaply, and how a center serving outside clients sits inside a university at all. Each has a bounded answer. The center, not the student, is the contracting party, so the obligations and the liability rest with the University, inside the structures it already uses for outside work, and the engagements are scoped and insured the same way. The client owns what is built, and we keep a teaching and research license to learn from it. Our students are paid and formed, their hours kept bounded, never treated as cheap labor. Settling these terms precisely is the first real work once the idea has your backing, and none of it is ground the University has not crossed before.

And the risk in taking it on is genuinely low. Across some forty of the world's leading university programs, every single piece of Forge is already running somewhere and working: open admission held to a hard bar, entry by built work, clinical-style supervision, real client engagements, students helping run the place. What no one has yet done is bring them together into one center that builds, deploys, owns what it ships, and is gradually run by the students who pass through it. Each piece is proven, so the risk is low. The combination is not, and I will not pretend otherwise: making the pieces hold together as one center is the real, unproven work, and some of it will only be learned by doing it. Being first to put the whole together is exactly the lead our school is positioned to take. The closest model at home is the UOB-SMU Asian Enterprise Institute, where students advise on real SME projects under industry mentors, genuine, well run, and a credit to the idea. Forge goes one step further: our students build, deploy, and maintain what they make, and in time come to run the center themselves. The full landscape is in the annex.

More for the Whole School

And a center like this gives back to the whole of our school, in more ways than money. Put plainly, Forge becomes the faculty's pipe to the real world: it grows the number of collaboration organizations, the real problem statements, and the data sources that the whole faculty can eventually draw on. It draws in standing relationships with industry that all of us can build on. It opens new lines of research and supplies the hard-won data to pursue them, the kind frontier labs keep to themselves, and the live problems it surfaces feed our own researchers and their centers upstream, a subset ready to graduate into their formal partnerships. It hands our newest colleagues something genuinely scarce: live problems, students ready to build, and grant work that already has real adoption behind it. It lifts our school's standing as the place where this is done. And it earns its own way. What Forge brings is added to what we have, not taken from it, and most of it lands on our school and our faculty directly, not only on our students. We do not have to divide the pie more thinly; we get to make it larger.

I will not set that against the cost, because the cost is real. In the early years Forge will take genuine hours of faculty supervision and give back nothing you can measure: no paper, no grant, only a student more capable than she was before. Those hours come out of the time of the very people whose time is already most fought over among us, and they come before the work has proven itself. I wish that part were not true. It is, and it deserves to be weighed exactly as written.

Beyond covering itself, Forge could in time do more. If it grows the way I hope it will, a surplus could help carry other innovation and entrepreneurship work across our school, rather than drawing on the same pool as everyone else. That is a direction to grow into, not a number I would promise you today.

Where the C-Innovators Sharpen Their Edge

Everything so far holds for every student who comes through Forge. There is a group for whom it holds with particular force. For our C-Innovators, Forge is exposure, experimentation, and the kind of inspiration that only real stakes can give. It is the proving ground for exactly the work the C-Innovators Programme asks of them across their three or four years: real businesses, real innovations, real problems carried all the way to a real result. The range and the sheer number of projects a C-Innovator can take on here, across many businesses and many technologies, build a depth of experience no comparable program can offer, because no comparable program has this much real work moving through it.

And this is where the deeper point about CIP comes in. The eventual goal of the C-Innovators Programme is that every C-Innovator starts their own business. That is where it is ultimately pointed, and building a good product is only one part of getting there. Doing it well takes hands-on experience across the many other aspects of a business: finding customers, managing client relationships, running operations, carrying the work all the way to real adoption. A classroom program, and product-building on its own, cannot give a student that. It has to be lived.

Forge adds exactly that dimension. Inside it, our C-Innovators do real client work, source their own SME clients and so build the business-development muscle first-hand, and grow into co-running the center itself. It is the full, hands-on experience of a business, gained while they are still students, and it is the practical preparation the Programme's goal actually requires. Inside Forge, our C-Innovators will have a real head start toward all of it: first choice of the projects that stretch them most, and first consideration for the roles that come with helping run the center. But a head start is not a finish line. Every position in Forge is earned by the work, the same way for everyone, and the door to it is open to every student. An open-cohort builder who ships more, and is trusted with more, rises ahead of a C-Innovator who does less; the advantage gives the C-Innovator the first move, not the last word. And it takes nothing from the open cohort. The opposite is true: that open cohort is what makes Forge work at all, the breadth of builders that lets us take on the volume of real work, and our C-Innovators sharpen their edge inside that larger body, earning their standing there rather than being handed it.

Because Forge is open to everyone, it also gives the Programme something rare: an interim way to see, in the open, the difference it makes. The students who are not in CIP are doing the same real work, on the same engagements, held to the same bar, so they are a natural point of comparison, and the difference the Programme makes shows up in how its students do alongside the rest, on real engagements rather than on a test. The reading has to be an honest one, and as educators we will read it honestly. Our CIP students are chosen, so a fair comparison has to account for where they began, not only where they arrive, and they carry a head start inside Forge as well, in the projects and the roles they grow into. Read with all of that in mind, it is a direction rather than a scoreboard, and it means something precisely because the students it measures against are real builders doing real work. It is an interim signal, while they are still training, of the thing that matters most: what the formation is genuinely adding.

Forge stands alongside the C-Innovators Programme, not beneath it. The Programme forms the kind of leader the years ahead will need, with breadth, judgment, and an ambition that reaches across technology, business, and society. Forge is where that capability is forged on real work, and made plain for everyone to see. The two belong together: one forms the person, the other shows what the person can do.

So what I am asking for is small, and it can be undone, and it is a decision only you can make. Back a single first engagement, and give me the mandate to run it. Not a budget, not a building, not a standing commitment, just one cohort, one client, watched closely, and stopped the moment it stops earning its keep. The supervision it will ask of us is real, and I have not hidden it. The rest of the case cannot be made on a page, and it would be wrong of me to pretend it could. It will be made the first time one of our students sits down with a real business, builds what it needs, and sees it through, in the open and on the record. That first step is small, and it is ours to take, and it is the whole of what I am asking for now.

What Other Schools Do - the Landscape

I surveyed more than forty leading universities. Every element of Forge already exists somewhere; the table below shows where, and where each precedent stops short.

Before I wrote this, I looked at how the world's leading universities promote innovation and entrepreneurship - more than forty institutions across the United States, the United Kingdom and Europe, the Nordics and the Baltics (Finland, Norway, and Estonia among them), Israel, Greater China, Korea and Japan, and this region, in both their computing and their business schools. The closest precedents to Forge are real and instructive - and each is scoped or selected in a way Forge is not.

Program	Where	What it does	Where it stops short of Forge
Real-client clinics
Data Science for Social Good	UChicago → CMU	Small teams + full-time mentors build for real government / non-profit partners	Scoped to data-for-good, not general SME software
Data Science Discovery	UC Berkeley	Credit-bearing, always-on clinic at scale (~250 projects and 900+ students a year)	Same data-only scope; not a general build clinic
Hack4Impact · Cal Blueprint	Penn · Berkeley	Students ship real, deployed software for non-profits ("deploy, not demo")	Non-profit-scoped; student-org, not a center with supervision + progression
"Software Project"	UT Tartu (Estonia)	Decade-plus general real-client software course; teams ship production software for real businesses, SMEs included (a bank's first mobile app, a startup's own product now in real use)	A one-semester faculty course, open enrollment (no building-test entry), judged on delivery rather than adoption, no maintain phase, unpaid - not a standing center
"Software Project"	Aalto · Helsinki (Finland)	Real-client software built for real firms over a semester; client pays ~EUR 3,000	Same one-semester-course limits; clients skew mid / large; no building-test, no adoption bar, no stipend, not AI-native
Customer-Driven Project (TDT4290)	NTNU (Norway)	Students build software for real external customers, for credit	One semester, no bar, judged on delivery rather than adoption - not a center
Problem-based learning · project semesters	Aalborg (Denmark)	Students take on real problems from real companies as coursework across the university (Aalborg's PBL), with project-based, company-collaborating software semesters (SDU, KEA)	Problem-based coursework rather than a dedicated software clinic; admission-gated and graded, unpaid, no building-test entry, judged on delivery rather than adoption - not a standing center
Consulting on real SME projects (the closest local model)
Asian Enterprise Institute · SME Consulting Program	UOB-SMU	Undergraduate teams take on real SME projects under external industry mentors (subsidized)	Students advise (reports, recommendations), they do not build or deploy; project-based, and they do not run the center
Entry test & verifiable record
Google Summer of Code · Outreachy	Open-source	Stipended real contributions to live open-source projects; selection is doing the work	Not a university; no SME client work
Creative Destruction Lab	U of Toronto	Holds a high bar throughout - culls between every session	Selective at entry; venture-focused, not open
Apprenticeship & work-integrated
Mayfield Fellows	Stanford	Paid, embedded startup apprenticeship (12 students/year)	Tiny and selective; sends students into others' startups
Overseas Colleges (NOC)	NUS	Paid startup internships abroad, for credit - Singapore's flagship learn-by-working model	Sends students out to others' startups; Forge keeps them in, owning real client work
Furnace	NUS SoC	In-school incubator (deferred-fee) - the in-school precedent	An incubator for ventures, not a teaching hospital doing supervised client work

Read region by region, a clear pattern falls out. Everyone has built something real and close to Forge, and everyone stops just short of it, each in their own instructive way.

The United States has the deepest and most diverse ecosystem of all. It runs data-for-good clinics at scale (the Data Science for Social Good family, Berkeley's Data Science Discovery), student build-orgs that ship genuinely deployed software rather than demos (Hack4Impact, Cal Blueprint), stipend-paying accelerators (MIT's delta v), and the federal I-Corps program that grew out of Stanford's Lean LaunchPad. It is a remarkable body of work. But its real-client clinics are scoped to data-for-good or non-profit software - none is a general software clinic that takes on real businesses.

The United Kingdom shapes its whole sector through incentive design: founder-favorable spinout equity (Oxford's standardized 80/20, Imperial's up-to-95 percent founder share) and deep-tech PhD venture programs (UCL's Conception X, Cambridge). It is venture- and IP-centered, and superbly run as such. What it does not have is an open real-client software clinic; the machinery points at spinouts, not at supervised delivery for outside firms.

Greater China resources innovation at a scale few can match: state- and corporate-backed, vertically integrated lab-to-spinout-to-holding-company-to-VC stacks (Tsinghua, PKU, HKUST), large university-committed capital, and some of the world's biggest makerspace and hardware infrastructure. It puts capital and space behind builders in a way that is genuinely formidable. What it resources, though, is capital and space, not supervised real-client delivery.

Korea and Japan each push the frontier in their own direction. Korea's KAIST OverEdge is the most explicitly AI-native accelerator anywhere - solo founders working alongside AI agents - and so the closest of all to Forge's "building is cheap" premise; yet it still resources seed capital and GPUs rather than judgment and proof. Japan's UTokyo scaled a non-dilutive, academia-led model into a national multi-university consortium, an impressive feat of reach. Both, though, are venture-creation engines, not client clinics.

Israel has the tightest research-to-commercialization coupling in the world: accelerators that run like funds tied straight to the IP pipeline (the Technion's DRIVE), university VCs (TAU Ventures), equity-free venture creation (HUJI's OPEN), and selective build-your-own-startup tracks (Reichman's Zell). Its one real-client analogue, the Technion's "Tech for Society", ships real software under supervision - but it is scoped to non-profits and hospitals, the social-impact carve-out rather than a general SME clinic, and its serious programs are pitch-selective startup funnels, the opposite of open entry.

Finland and Norway come the closest anyone does to the clinic function itself: real-client software courses where students build for real businesses under faculty supervision. Finland's Aalto and Helsinki run a "Software Project" for a real client, a subsidized partner fee and all; Norway's NTNU has students build for real external customers through its Customer-Driven Project course (TDT4290). These are genuinely real, and genuinely close. But each is a one-semester course: open by enrollment with no sustained bar, judged on delivery rather than on whether the business adopts what was built, ending at handover with no maintenance, unpaid, and never a standing center. (Finland's clients skew mid-sized and large, not SMEs.)

Denmark reaches real company work by a different route. Aalborg's problem-based learning puts students onto real problems from real companies as coursework across the whole university, and there are project-based software semesters at SDU and KEA built around collaborating with real firms. That is real company work embedded in the degree, and it deserves the credit. But it is problem-based coursework rather than a standing clinic: admission-gated, graded, unpaid, and it stops where the course does, without the sustained bar, the adoption test, or the maintained deployment. It is a partial real-client case rather than a software clinic outright - closer to Finland's in spirit than the two are alike, and a step short of even that.

The closest local comparison is the UOB-SMU Asian Enterprise Institute's SME Consulting Program - undergraduate teams taking on real SME projects under external industry mentors. It is a good model, and it marks exactly where Forge goes further. There, students advise: they study a business and hand over recommendations. In Forge they build: they ship and deploy working software, and own whether it actually works in the business. Their work is guided by industry mentors alone; Forge blends practitioners with faculty, whose involvement also turns each engagement into material for their own research. And there, students are project-based visitors; in Forge, the senior ones help run the center itself.

Reading the full dataset, four positions are conspicuously empty - and they are exactly the ones Forge takes:

• Open access held to a high bar. Programs are open-and-easy or selective-at-entry; almost none stay open to all yet hold a hard bar throughout via continuous public proof.
• Entry by a building test, not a pitch. No university I could find gates entry on both a deployed build and a merged open-source contribution.
• The general software clinic run as a standing teaching hospital. The general real-client software clinic does exist - as a one-semester course (Tartu, Aalto, and others do it, SMEs included). What no one runs is that clinic as a standing teaching hospital: building-test-gated entry, advancement judged by real adoption, a maintain phase, self-funding on the value it creates, AI-native resourcing, and the door held open at a high bar throughout a career rather than for a single term.
• Resourcing judgment, not capital. Even the 2024-26 AI wave still resources building (credits, GPUs, funds); almost nothing resources judgment, experience, and proof.

And then there is Estonia, which is worth ending on, because it is the closest national analogue to Singapore of them all. Estonia is small, highly tech-literate, and built on a mature digital-government stack with an explicit SME-digitalization agenda - much like us. It has also run the thing longest: the University of Tartu's "Software Project" has, for over a decade, put student teams onto real production software for real businesses, SMEs included - a bank's first mobile app and a startup's own product now in real use among them. So Estonia has already proved the part everyone doubts: that SMEs will bring real problems to a university-brokered program, and that students can deliver on them. What Estonia never built is the institution around it - the earned way in, the public verifiable record that confers standing, the economics that let it sustain itself and grow. It has the clinic; it never made it a teaching hospital. That gap - not whether the work can be done, but whether it can be made into a durable, high-standing, self-funding institution - is precisely the opening Forge is built to take, in the country whose situation most resembles Estonia's own.

Full methodology and the institution-by-institution dataset are available on request (working files: the landscape survey of more than forty institutions across computing and business schools, US / UK / Europe / Nordics / Baltics / Israel / Greater China / Korea / Japan / Singapore).