We’re live on Product Hunt today — your upvote would make our day 🙌
Live on Product Hunt today →
We’re live on Product Hunt today — your upvote would make our day 🙌
Live on Product Hunt today →
This report was generated by our Deep Research agent and may contain mistakes.
Did we get something wrong? DM @oscrhong and we'll fix it ASAP!
Upgrade to Pro to get implementation-ready specs for every company, the full report library, and 5 on-demand report requests per month.
Make School launched in 2012 as MakeGamesWithUs, a Y Combinator-backed iOS game publishing platform and CS summer camp.By 2014, it had pivoted into something far more ambitious: a two-year, project-based alternative to a four-year computer science degree, financed through income share agreements (ISAs) that let students pay nothing upfront and share income after graduation.
The company raised $30 million from credible investors, achieved a landmark accreditation partnership with Dominican University of California, and enrolled hundreds of students from underrepresented backgrounds.It collapsed in July 2021 under the simultaneous weight of a denied independent accreditation application, a lawsuit from 47 former students alleging predatory ISA terms, and an insolvency proceeding.
The core thesis of failure: Make School's ISA-based business model—designed to align incentives between students and the institution—became a regulatory and legal liability that exposed predatory lending practices, while a simultaneous bet on independent accreditation failed, leaving the company financially insolvent and institutionally illegitimate at the same moment.
Jeremy Rossmann and Ashutosh (Ashu) Desai met as students at Menlo School in Atherton, California, where they bonded over computer science. Both went on to elite universities—Rossmann to MIT, Desai to UCLA—before dropping out to pursue their startup. The decision to leave school was not impulsive. It was contingent on a specific external validation. [1]
"We only made the decision to drop out once we got into YC," Desai said. "YC acted as this credential that gave us access to this Silicon Valley network." [2]
The founding irony is hard to miss: two founders who needed an institutional credential to justify leaving an institution would spend the next decade arguing that institutional credentials were overrated.
In 2012, the company they built through YC was modest in scope. MakeGamesWithUs offered CS summer camps for high school and college students and operated as an iOS game publishing platform, helping young developers build and ship games to the App Store. [3] The product was real, the market was small, and the mission was narrow.
The 2014 pivot changed everything. The company rebranded as Make School and launched a gap year program, signaling a fundamental shift in ambition: from a game-publishing platform to a full alternative to higher education. [4] The scope expansion was dramatic. Building a summer camp requires a curriculum and some instructors. Building an accredited degree-granting institution requires navigating state licensing, federal financial aid rules, regional accreditation bodies, and consumer protection law—a regulatory surface area the founders appear to have underestimated.
The mission that animated the pivot was explicitly equity-driven. "Our purpose is to create avenues of upward mobility for students of all backgrounds interested in science and technology," Desai said. [5] Rossmann framed the product vision in terms of student agency: "We founded Make School to empower students to build and ship products. Creating a product that improves the lives of others is a thrilling and fulfilling experience that opens the door to a successful career in tech." [6]
Both statements reflect genuine conviction. Neither anticipated the regulatory collision course the company was already on.
Make School's core product was a two-year, full-time applied computer science program based in San Francisco. It was designed to replace the four-year bachelor's degree with something faster, cheaper, and more directly connected to industry employment.
The Curriculum
The program was project-based and product-focused. Rather than teaching computer science through lectures and problem sets, Make School required students to build working software applications at each stage of the curriculum. [24] Students chose one of four concentrations: Frontend Web Development, Backend Web Development, Mobile Development, or Data Science. [25] The curriculum was developed in partnership with industry employers including Yelp, Microsoft, and Lyft, giving it direct relevance to the hiring practices of the companies students hoped to join. [26]
Rossmann articulated the philosophy behind the curriculum's constant evolution: "Our core philosophy is if you teach the same thing two years in a row, it's got to be wrong because computer science as a field and software engineering as a discipline is moving so fast." [27]
The Two-Year Compression
Eliminating the four-year structure was not just a marketing decision. It removed "summer melt"—the phenomenon where students who enroll in September fail to return after summer breaks—and significantly reduced the total cost of tuition by cutting two years of instruction and living expenses. [28] Students attended year-round, completing in two calendar years what traditional programs spread across four.
The ISA Model
The financial engine of the entire enterprise was the income share agreement. Students paid nothing upfront. After graduation, they paid a percentage of their income for a defined period. Ninety percent of Make School students opted into ISAs. Two in three also rolled living expenses—housing, food—into the same ISA arrangement, meaning their total financial obligation extended well beyond tuition. [29] ISAs were originated and serviced through a partnership with Vemo Education.
The ISA model was Make School's most distinctive feature and, ultimately, its most dangerous one. The theoretical alignment of incentives—the school only profits if graduates get good jobs—was real. The practical implementation, as the Student Borrower Protection Center later documented, could trap students in obligations totaling up to $250,000 over a decade. [30]
The Accreditation Partnership
In November 2018, Make School achieved something genuinely unprecedented: it became the first post-secondary program accredited under WSCUC's incubation policy, operating in partnership with Dominican University of California. [31] Under the arrangement, Dominican faculty taught general education courses on Make School's campus, and Make School faculty taught computer science courses at Dominican. Students received a Bachelor of Science in Applied Computer Science with Dominican's name on the degree. [32]
This was a genuine regulatory breakthrough. It also created a structural dependency that would prove fatal: Make School's institutional legitimacy was not its own. It was borrowed from Dominican.
Make School's primary target was students who wanted careers in software engineering but found traditional four-year universities either too expensive, too slow, or too disconnected from industry practice. The school specifically recruited students from underrepresented backgrounds. As of the 2018 accreditation announcement, approximately 45 percent of enrolled students were students of color. [33]
This demographic profile was central to the equity mission Desai articulated publicly. It also meant the student population was disproportionately low-income—and therefore disproportionately vulnerable to the financial terms embedded in the ISA contracts they were signing.
The addressable market for alternative CS education was large and growing during Make School's operating years. U.S. coding bootcamp enrollment grew from roughly 6,000 students in 2013 to over 23,000 by 2019, according to industry surveys. The broader market for post-secondary CS education encompassed millions of students annually. Make School was not competing for the entire market—it was competing for a specific segment: students who wanted a degree credential (not just a bootcamp certificate), were willing to commit to two full-time years, and preferred ISA financing over traditional student loans.
Venrock partner Tom Willerer framed the opportunity at the Series B: "Make School is tackling the three big issues in higher education: cost, relevance, and equity." [34] The framing was accurate. The execution was not.
Make School competed across three distinct categories simultaneously, which complicated its positioning.
Against traditional four-year CS programs at universities, Make School offered speed (two years vs. four), lower upfront cost (ISA vs. tuition), and industry-aligned curriculum. The disadvantage was brand recognition: a Dominican University degree in Applied Computer Science carried less weight with employers than a degree from a UC campus or a private university with an established CS reputation.
Against coding bootcamps—General Assembly, App Academy, Flatiron School, Lambda School—Make School offered an accredited degree credential rather than a certificate, a longer and more rigorous program, and a broader curriculum. Lambda School (later Bloom Institute of Technology) was the most direct competitor, also using an ISA model and also facing regulatory scrutiny for its ISA terms during the same period.
Against community colleges and state schools offering two-year CS programs, Make School offered a more intensive, industry-connected curriculum and a San Francisco location that facilitated internships and networking. The disadvantage was cost: even with ISAs, Make School's total financial obligation could far exceed what a community college student would pay.
The competitive position was genuinely differentiated. No other institution in 2018 offered an accredited two-year applied CS degree with ISA financing in San Francisco. The differentiation was real. The regulatory and financial structure supporting it was not.
Make School operated as a Public Benefit Corporation (PBC) and generated revenue primarily through income share agreements. Students paid nothing upfront; after graduation, they paid a percentage of their income for a defined period. ISAs were structured and serviced through a partnership with Vemo Education. With 90 percent of students opting into ISAs and two-thirds rolling living expenses into the same arrangement, [35] the company's revenue was entirely deferred—dependent on graduates finding well-paying jobs and making payments over time.
This created a fundamental cash flow problem: Make School had to fund operations today against income it would not collect for years. Venture capital filled the gap. The $30 million raised across multiple rounds [36] was effectively subsidizing current operations against future ISA collections. When the ISA program was cancelled in July 2021, the revenue model collapsed entirely—and with it, the company.
In December 2020, Make School announced a move toward nonprofit status, citing the goal of seeking grants and donations to help low-income students cover living expenses. [37] The pivot away from ISAs signaled that the founders recognized the model was untenable before the formal collapse.
Make School's enrollment grew meaningfully in its early years. The program had 40 students enrolled in 2017. By the 2018 accreditation announcement, that number had grown to 110. [38] By April 2019, 202 students had enrolled since the program launched in 2014. [39]
The completion rate was alarming. As of April 2019, only 3 of those 202 enrolled students had completed the two-year program, with roughly half still currently enrolled. [40] The program was young enough that this figure may partly reflect the timeline—students who enrolled in 2018 could not have graduated by April 2019. But the number was still strikingly low for a company that had raised $30 million and was planning to scale to 1,000 students.
Graduate outcomes, as reported by the founders, were strong. Desai cited 66 graduates earning an average annual salary of $95,000. [41] Students had interned and worked at 60 companies including Apple, Tesla, NASA, PayPal, Twitter, and Dropbox. [42] These figures came from the founder and were not independently verified. The SBPC later alleged that Make School misrepresented earnings data to prospective students, including claiming graduates would earn starting salaries of $100,000. [43]
The investor base was credible: Venrock, Kapor Capital, Learn Capital, Tim Draper, Alexis Ohanian, and Initialized Capital all participated across multiple rounds. [44] Their participation validated the model externally. It did not resolve the regulatory exposure that was already accumulating.
Make School's collapse was not a single event. It was the convergence of three compounding failures: a regulatory violation the company allegedly concealed and ignored, a business model that generated predatory financial obligations for the students it claimed to serve, and an institutional legitimacy that was borrowed rather than owned. All three came due simultaneously in July 2021.
The most damaging fact in the Make School record is not the lawsuit or the denied accreditation. It is what allegedly happened in 2018 and what the company did—or did not do—in response.
From 2016 to 2018, Make School operated unlawfully as a for-profit college in California. [45] In 2018, a California regulator issued an order requiring the school to refund all money paid by students and to cancel the ISAs it had engineered with Vemo Education. [46]
According to the July 2021 lawsuit filed by 47 former students, Make School did not comply with that order. Instead, the lawsuit alleged, the company actively concealed the order from students and continued to enroll new students after it was issued. [47] Vemo allegedly continued to collect on ISAs that the regulatory order had rendered unenforceable.
This sequence—if the allegations are accurate—transformed a compliance failure into potential fraud. A school that unknowingly violated state licensing law and then corrected course is a cautionary tale. A school that received a regulatory order, concealed it, and continued enrolling students is something else. The distinction matters legally and reputationally.
The company's attempt to address the underlying licensing problem was the November 2018 WSCUC accreditation partnership with Dominican. That partnership did resolve the accreditation gap going forward. It did not retroactively cure the 2016–2018 unlawful operation, and it did not address the students who had enrolled during that period.
The income share agreement was Make School's most innovative feature and its most consequential liability. The theoretical design was sound: students pay nothing upfront, the school profits only if graduates find good jobs, incentives align. The practical implementation, as documented by the SBPC, produced obligations that could total $250,000 over a decade. [48]
For context: the average total cost of a four-year degree at a public university in the United States, including tuition, fees, and room and board, was approximately $100,000 during Make School's operating years. An ISA that could generate $250,000 in obligations made Make School more expensive than most traditional four-year universities—directly contradicting the school's core value proposition.
The structural problem was identified clearly by a former student: "Not every ISA is predatory. It's just that there's an overwhelming temptation from these schools to get to this place that's predatory." [49] The SBPC's Managing Counsel Mike Pierce was less charitable: the ISAs "were sold as shiny new financial innovations but, as today's lawsuit makes clear, they were nothing more than plain old predatory lending." [50]
The compounding factor was the living expense ISA. Two-thirds of students rolled housing and food costs into the same ISA arrangement. [51] This meant students were not just financing tuition—they were financing their entire cost of living in San Francisco, one of the most expensive cities in the United States, through an instrument that could generate obligations far exceeding the original amounts borrowed.
Make School's attempt to address the ISA problem came in December 2020, when the company announced it was moving away from ISAs and toward nonprofit status. [52] The pivot was too late. The lawsuit was filed seven months later, and the ISA program was cancelled as part of the insolvency proceeding—not as a voluntary reform.
Desai called accreditation "really a lynchpin for us. It feels so much more tangible and real, and it's a start of something that will be a lasting institution." [53] He was right about the importance. He was wrong about the durability.
The WSCUC incubation partnership with Dominican was a genuine regulatory achievement. It was also a structural dependency. Make School's ability to grant degrees was contingent on Dominican's continued participation and WSCUC's continued approval. The company had no independent accreditation of its own.
In July 2021, WSCUC denied Make School's application for independent accreditation. [54] The specific reasons for the denial are not publicly documented—whether it was the ongoing lawsuit, the financial instability, the regulatory violations, or curriculum deficiencies is unknown. The effect was unambiguous: without independent accreditation, Make School could not operate as a standalone degree-granting institution. The Dominican partnership could not substitute for that standing indefinitely, and Dominican was not obligated to continue the arrangement once the for-profit entity became insolvent.
The timing was catastrophic. The accreditation denial and the insolvency proceeding were announced on the same day—July 1, 2021. Rossmann's email to stakeholders confirmed that Make School PBC had "started an insolvency proceeding" known as an Assignment for Benefit of Creditors and was "in the process of shutting down." [55] He stated: "Make School PBC has now handed over all operations of the bachelor's program to a nonprofit, has cancelled its ISA program, and is winding down. The college will continue to exist, but under new management and without ISAs." [56]
The July 16, 2021 student town hall, titled "End Game," captured the human dimension of the collapse. A student on the Zoom call said: "We applied to something that is no longer in this room. We applied to something that no longer exists." [57]
When Make School closed, 167 students were left mid-program. Dominican absorbed the wreckage, attempting to help students continue their education. The majority left without any credential to show for their time and financial commitment. [58]
The SBPC found that Make School and Dominican had driven hundreds of largely low-income students to take on thousands of dollars of predatory private student loans, through promises of high-paying tech careers. [59] The school had also allegedly misrepresented its status as a future "yellow ribbon" school eligible for GI benefits—an approval it reportedly never obtained. [60]
The deepest tension in the Make School story is between its stated mission and its business model. The school explicitly recruited low-income students of color, citing equity as its core value. [61] It then financed their education through an instrument that could generate $250,000 in obligations. The students most harmed by the ISA terms were the students the school claimed to be serving.
This is not a contradiction that can be explained by bad luck or regulatory surprise. The ISA terms were known. The student demographics were known. The combination was a structural problem embedded in the business model from the beginning.
Borrowed accreditation is not institutional legitimacy. Make School's partnership with Dominican University was a creative regulatory solution, but it created a dependency that the company could not survive losing. Any institution whose degree-granting authority depends on a third party's continued participation is one relationship away from institutional collapse. The lesson for edtech operators: accreditation through partnership is a bridge, not a destination. Independent accreditation must be the goal, and the path to it must be financially and legally clean.
ISA structures require regulatory clarity before enrollment, not after. Make School operated unlawfully as a for-profit college in California from 2016 to 2018 and received a regulatory order it allegedly ignored. The ISA model was novel enough that regulatory frameworks were still developing—but novelty is not a defense against consumer protection law. Companies deploying ISAs must obtain explicit regulatory approval in each jurisdiction before enrolling students, not after receiving a cease-and-desist. Lambda School (Bloom Institute of Technology) faced similar regulatory scrutiny during the same period, suggesting the problem was industry-wide, not idiosyncratic to Make School.
Serving low-income students with deferred-payment instruments creates structural ethical risk. The ISA model's incentive alignment—the school profits only if graduates succeed—is theoretically sound. In practice, the temptation to maximize the financial obligation extracted from students is overwhelming, as a former student identified explicitly. When the student population is disproportionately low-income and financially unsophisticated, the power asymmetry between institution and student makes predatory outcomes likely even without predatory intent. Mission-driven institutions serving vulnerable populations need external constraints on their financial instruments, not just internal good intentions.
Completion rates are a leading indicator of institutional health. As of April 2019, only 3 of 202 enrolled students had completed Make School's two-year program. [62] This figure was available to investors at the Series B and did not prevent the $15 million raise. Completion rates in alternative education are a direct measure of whether the product works. A 1.5 percent completion rate at the time of a major fundraise was a warning sign that the program's design, student support, or financial structure was failing students—and that the ISA revenue model was therefore built on a fragile foundation.
The Dominican partnership illustrates a systemic risk in higher education branding. Established universities that lend their accreditation brand to for-profit operators without adequate oversight expose their students, their reputations, and their regulatory standing to the operator's conduct. Dominican's involvement gave Make School legitimacy it could not have obtained independently. When Make School collapsed, Dominican was left managing 167 displaced students and facing SBPC scrutiny. The lesson for universities considering similar partnerships: the reputational and regulatory liability of a failed partnership can far exceed the revenue or mission benefit of a successful one.
Ready to rebuild Make School?
Implementation-ready specs, every report, and 5 on-demand requests each month.