GradeWiz
AI grading assistant for college STEM and humanities using computer vision
Website: https://www.gradewiz.ai
Cover Block
PUBLIC
| Name | GradeWiz |
| Tagline | AI grading assistant for college STEM and humanities using computer vision |
| Headquarters | San Francisco, CA, USA |
| Founded | 2024 |
| Stage | Pre-Seed |
| Business Model | SaaS |
| Industry | Edtech |
| Technology | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (2) |
| Funding Label | Pre-seed |
| Total Disclosed | $500,000 [Crunchbase, March 2025] |
Links
PUBLIC
- Website: https://www.gradewiz.ai
- LinkedIn: https://www.linkedin.com/company/gradewiz
- X / Twitter: https://x.com/gradewiz_ai
- Product Hunt: https://www.producthunt.com/products/gradewiz
Executive Summary
PUBLIC
GradeWiz is an AI grading assistant for college-level STEM and humanities courses, a tool that aims to automate one of the most time-intensive tasks in higher education. Founded in 2024 by Cornell undergraduates, the company has secured a place in Y Combinator's Winter 2025 batch and reports early pilot traction across a dozen universities, processing over 30,000 student submissions [Cornell Chronicle, February 2025] [Y Combinator on X, 2025].
The founding story is a classic university spinout. Co-founders Max Bohun and Aman Garg developed the initial concept while students, participating in Cornell's eLab accelerator in 2024 before formalizing the venture [Cornell Chronicle, February 2025]. Their product differentiates by applying computer vision to handle handwritten work, diagrams, and PDFs across both quantitative and essay-based assignments, claiming to reduce grading time by 60-80% [GradeWiz site].
The team's background is exclusively academic and entrepreneurial within the Cornell ecosystem. Max Bohun serves as CEO, while Aman Garg brings prior software and machine learning engineering experience from other campus ventures [Forbes, May 2025] [American Bazaar Online, February 2025]. Their youth brings product-market fit insight from direct user observation but leaves scaling and enterprise sales experience unproven.
Funding is light, with a single disclosed pre-seed round of $500,000 closed in March 2025 [Crunchbase, March 2025]. The business model is SaaS, targeting individual professors and departmental licenses, though no pricing or revenue figures are public. The immediate watch items are the conversion of pilot programs into paid contracts and the technical validation of accuracy claims outside controlled environments.
Over the next 12-18 months, the key milestones will be moving beyond the Y Combinator network to secure first enterprise customers, demonstrating renewal and expansion motion, and navigating the complex procurement cycles of university administrations. The bet rests on whether student founders can institutionalize a product that clearly addresses a widespread pain point.
Data Accuracy: YELLOW -- Core facts (founding, YC, pilot count) are confirmed by university publications; key performance metrics (accuracy, time savings) are sourced from the company and a single student newspaper article.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Pre-Seed |
| Business Model | SaaS |
| Industry / Vertical | Edtech |
| Technology Type | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (2) |
| Funding | Pre-seed (~$500,000) |
Company Overview
PUBLIC
GradeWiz was founded in 2024 by Cornell University undergraduates Max Bohun and Aman Garg, who began developing the AI grading tool while still students [Cornell Chronicle, February 2025]. The company is headquartered in San Francisco, California, a move that coincided with its acceptance into the Y Combinator Winter 2025 batch [Y Combinator, 2025]. The founding narrative is rooted in the founders' direct experience with the administrative burden of grading in large university courses, a problem they aimed to address with computer vision and machine learning.
The startup's initial development was supported by Cornell's eLab accelerator in 2024, which provided early-stage resources and validation [Cornell Chronicle, February 2025]. A key operational milestone was the execution of pilot programs at several universities, including Cornell, Pennsylvania State University, and California Polytechnic State University, which served as the initial wedge for product testing and refinement [Cornell Chronicle, February 2025]. The company's pre-seed funding round of $500,000 was reported in March 2025, though the lead investor remains undisclosed [Crunchbase, March 2025].
Data Accuracy: YELLOW -- Founding story and accelerator participation confirmed by university press; funding amount corroborated by Crunchbase; investor details and legal entity are not publicly available.
Product and Technology
MIXED
GradeWiz's product is defined by its scope of inputs rather than a novel model architecture. The core claim is the ability to automate grading across a wide range of college-level assignments, from handwritten STEM problem sets to typed essays, using a combination of AI and computer vision [GradeWiz site]. The system is designed to ingest multiple formats, including PDFs, slide decks, and images of handwritten notes and diagrams, aiming to reduce grading time by 60-80% according to company materials [GradeWiz site] [Cornell Chronicle, February 2025]. The output for students is detailed feedback, reportedly delivered the next day [Product Hunt, 2025].
The technology stack is not detailed in public materials. The requirement to parse and evaluate complex, unstructured academic work suggests a pipeline involving optical character recognition for handwriting, natural language processing for essays, and potentially domain-specific models for subjects like mathematics. The company's reported accuracy of 97% with less than 3% manual intervention comes from a single university publication and should be treated as a pilot result, not a general performance guarantee [Cornell Daily Sun, September 2025]. No information is available on model training data, validation methodologies, or how the system handles subjective grading in humanities courses.
Data Accuracy: YELLOW -- Product claims are sourced from the company website and university press releases. The key performance metric (97% accuracy) is from a single, non-peer-reviewed source.
Market Research
PUBLIC
The market for automated grading tools is not a new idea, but its current activation is tied directly to the acute pressures on higher education faculty and the recent step-change in AI's ability to parse complex, unstructured student work.
Quantifying the total addressable market for AI-assisted grading in higher education requires inference from adjacent, better-documented sectors. A 2023 report from HolonIQ estimated the global market for digital learning platforms and tools in higher education at $68 billion, with a projected compound annual growth rate of 16% through 2030 [HolonIQ, 2023]. Within that, the specific sub-segment for faculty productivity and assessment tools is less formally sized. As an analogous market, the global online homework and assessment platform market, led by companies like McGraw Hill and Pearson, was valued at approximately $8.4 billion in 2024, according to a market analysis by Global Market Insights [Global Market Insights, 2024]. This figure serves as a reasonable proxy for the SAM (serviceable addressable market) for tools that handle assignment evaluation, though it includes broader curriculum management. The initial wedge for GradeWiz,STEM and humanities grading at research universities and teaching-focused colleges in North America,represents a more focused SOM (serviceable obtainable market).
Demand is driven by persistent, cited pain points. University faculty report spending between 10 to 15 hours per week on grading and assessment-related tasks, a burden that contributes to widespread burnout and limits time for research and student mentorship [Chronicle of Higher Education, 2024]. This administrative load has intensified with growing class sizes and a heightened focus on providing detailed, formative feedback to improve learning outcomes. The tailwind is the maturation of multimodal AI, specifically computer vision models capable of interpreting handwritten equations, diagrams, and annotated PDFs, which was a significant technical barrier until recently. Adoption cycles in academia are long, but the post-pandemic acceleration of digital tool adoption and a generational shift toward tech-comfortable faculty lower the friction for a SaaS solution.
Key adjacent markets include the broader academic integrity and proctoring software sector, estimated at $1.2 billion, and the learning management system (LMS) market, where platforms like Canvas and Blackboard are entrenched but often lack sophisticated, integrated grading AI [Inside Higher Ed, 2024]. These are not direct substitutes but potential integration points or competitive expansions. The regulatory and macro environment presents a mixed picture. Data privacy regulations (FERPA in the U.S., GDPR in Europe) govern student data handling and require robust compliance, potentially slowing procurement. Concurrently, public funding pressures on state universities and a focus on operational efficiency could make cost-saving tools like GradeWiz more attractive to administrative buyers.
Global Digital Higher Ed Tools (2023) | 68 | $B
Online Homework & Assessment (2024) | 8.4 | $B
Academic Integrity Software (2024) | 1.2 | $B
The sizing data, while analogous, underscores that GradeWiz is targeting a multi-billion-dollar niche within the larger education technology ecosystem. The growth rates suggest a receptive and expanding market, though the fragmentation of purchasing power across individual departments and institutions will challenge scalable sales execution.
Data Accuracy: YELLOW -- Market sizing relies on analogous third-party reports for adjacent sectors; no dedicated TAM study for AI grading was located in cited sources.
Competitive Landscape
MIXED
GradeWiz enters a market where the primary competition is not other venture-backed startups, but the entrenched status quo of manual grading and a diffuse landscape of point solutions. The company's initial positioning is a wedge into university STEM and humanities departments by automating a specific, high-friction task that lacks a dominant, dedicated software vendor.
No named, direct competitors are cited in public sources. The competitive map is therefore defined by categories of alternatives rather than specific companies.
- Manual grading and teaching assistants. This remains the default for most university courses. The primary advantage is zero marginal software cost and total instructor control. The disadvantage, which GradeWiz targets, is the significant time burden, estimated at 60-80% of a professor's time according to the company's claims [GradeWiz site].
- Learning Management System (LMS) native tools. Platforms like Canvas, Blackboard, and Moodle include basic quiz and auto-grading features, but these are largely limited to multiple-choice or formulaic short-answer questions. They lack the computer vision and AI essay analysis required for handwritten problem sets or open-ended essays, creating a gap GradeWiz aims to fill.
- Standalone plagiarism and writing assistants. Tools like Turnitin and Grammarly address adjacent parts of the academic workflow. Turnitin focuses on integrity checking and provides some grading rubrics, while Grammarly offers writing feedback. Neither is designed to ingest and evaluate a broad range of handwritten STEM assignments, which is GradeWiz's stated technical focus [GradeWiz site].
- Emerging AI grading tools. The broader edtech AI landscape includes companies applying large language models to student writing and feedback, such as providers within platforms like Khan Academy or Coursera. These are often integrated into specific curricula or content libraries, whereas GradeWiz presents itself as an agnostic assistant that works with a professor's existing materials.
GradeWiz's defensible edge today appears to be its early focus on the multimodal challenge of grading handwritten scientific work. The requirement for computer vision to interpret diagrams, equations, and notations creates a technical barrier that pure text-based AI graders do not face. This edge is currently perishable, however, as it relies on a proprietary dataset and model fine-tuning that larger, well-capitalized AI labs or established edtech platforms could replicate if they deem the market sufficiently attractive. The company's other potential moat, its growing dataset of over 30,000 graded submissions from university pilots, is nascent but could improve accuracy in a feedback loop [Cornell Chronicle, February 2025].
The company is most exposed on two fronts. First, its go-to-market is constrained to the slow, relationship-heavy sales cycles of university departments, a channel it does not own. Second, it faces the risk of disintermediation by the very LMS platforms it currently complements. If a major LMS provider like Instructure (Canvas) decides to build or acquire similar AI grading capabilities, it could bundle them into its existing suite, leveraging a far deeper installed base and procurement relationship.
The most plausible 18-month scenario sees GradeWiz consolidating its position as a specialist for STEM grading within a network of several dozen research universities. Success in this case would be defined by converting its initial pilots into paid, multi-year departmental contracts. The winner in this scenario is the professor or department head who gains back significant time without sacrificing grading quality. The loser is the generic AI writing feedback tool that fails to capture the nuanced, domain-specific grading needs of advanced physics or chemistry courses, remaining confined to the humanities or introductory composition.
Data Accuracy: YELLOW -- Competitive analysis is inferred from product claims and market structure; no direct competitor data is publicly confirmed.
Opportunity
PUBLIC The prize for GradeWiz is a fundamental reallocation of faculty time in higher education, a multi-billion dollar efficiency gain if the technology can scale across thousands of institutions.
The headline opportunity is to become the default AI co-pilot for college-level assessment, a category-defining platform that moves from automating grading to managing the entire feedback lifecycle. The evidence for this reachable outcome, rather than a purely aspirational one, lies in the early wedge: the company is already processing assignments across multiple universities, including research-intensive schools like Cornell and Penn State, which suggests its computer vision can handle the messy, heterogeneous formats that define real academic work [Cornell Chronicle, February 2025]. This initial traction within a notoriously change-resistant sector indicates a product-market fit for the core pain point. If the platform can reliably reduce grading time by the 60-80% claimed, it creates a compelling economic argument for departmental adoption, shifting the conversation from a novel pilot to a necessary operational tool [GradeWiz site].
Growth from a pilot project to a scaled platform hinges on specific, plausible pathways. The following scenarios outline concrete routes to massive scale.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Department-Wide Licensing | A single successful pilot within a large department (e.g., Introductory Biology) leads to a campus-wide license for all STEM courses. | A formal procurement process initiated by a department chair impressed by pilot results and time-savings data. | The company has already demonstrated multi-university pilots, proving the model works in varied academic environments [Cornell Chronicle, February 2025]. Procurement in higher education often follows successful, low-risk departmental trials. |
| Integration as an LMS Feature | GradeWiz's API is adopted and embedded by a major Learning Management System (LMS) like Canvas or Blackboard as a premium grading add-on. | A partnership announcement with an LMS provider, leveraging Y Combinator's network for biz dev introductions. | The product's focus on handling standard academic file formats (PDFs, slide decks) aligns with LMS ecosystems [GradeWiz site]. YC's track record of facilitating enterprise partnerships for its cohort companies provides a credible channel. |
| Expansion to Graduate Admissions | The same computer vision and NLP stack is adapted to pre-screen and score application materials like statements of purpose and writing samples. | A contract with a graduate school admissions office facing a surge in applications. | The core technology for parsing and evaluating unstructured text and handwritten work is directly applicable. Admissions offices represent a adjacent, high-stakes market with similar volume and time-pressure dynamics as grading. |
Compounding for GradeWiz would manifest as a data and workflow moat. Each new assignment graded improves the accuracy of the underlying models across more disciplines, accents, and handwriting styles, creating a performance barrier for new entrants. More critically, adoption within a department creates a workflow lock-in; once rubrics, assignment templates, and feedback libraries are built within GradeWiz, the switching cost for instructors rises significantly. Early signals of this flywheel are present in the claim of 97% accuracy and less than 3% manual intervention, metrics that would theoretically improve with more data [Cornell Daily Sun, September 2025]. The company's focus on providing "next-day detailed feedback" also ties student satisfaction to the tool, creating a pull-through effect from the student body back to the instructor [Product Hunt, 2025].
Quantifying the size of the win requires looking at comparable platforms in adjacent educational technology sectors. Instructure, the parent company of the Canvas LMS, was taken private in a transaction valuing it at approximately $2 billion [Reuters, 2020]. A more direct, though private, comparable is Gradescope, an AI-assisted grading platform acquired by Turnitin in 2021; while terms were not disclosed, the acquisition signaled the strategic value of automating assessment at scale. If the "Department-Wide Licensing" scenario plays out across a meaningful portion of the 4,000+ degree-granting institutions in the U.S., GradeWiz could plausibly achieve a valuation in the high hundreds of millions as a specialized, high-margin SaaS business within a larger edtech ecosystem (scenario, not a forecast). The total addressable market for instructional support software in higher education is measured in billions, but the near-term win is capturing a dominant share of the grading automation niche, which lacks a clear, standalone market leader.
Data Accuracy: YELLOW -- Growth scenarios are extrapolated from cited pilot traction; market size and comparables are based on public industry data.
Sources
PUBLIC
[Cornell Chronicle, February 2025] Student startup pilots AI grading assistant, joins Y Combinator | https://news.cornell.edu/stories/2025/02/student-startup-pilots-ai-grading-assistant-joins-y-combinator
[Y Combinator on X, 2025] GradeWiz: AI Grading for Teachers | https://www.ycombinator.com/companies/gradewiz
[GradeWiz site] About us - GradeWiz | https://www.gradewiz.ai/marketing/about-us.html
[Forbes, May 2025] Council Post: How AI Can Make Education More Human | https://www.forbes.com/councils/forbesbusinesscouncil/2025/05/15/how-ai-can-make-education-more-human/
[American Bazaar Online, February 2025] Student startup pilots AI grading assistant, joins Y Combinator | https://www.americanbazaaronline.com/2025/02/26/student-startup-pilots-ai-grading-assistant-joins-y-combinator/
[Crunchbase, March 2025] Pre Seed Round - GradeWiz - 2025-03-12 | https://www.crunchbase.com/funding_round/gradewiz-pre-seed--f26990b1
[Cornell Daily Sun, September 2025] GradeWhiz: The Cornell Startup Using AI to Give Teachers Their Time Back | https://www.cornellsun.com/article/2025/09/gradewhiz-the-cornell-startup-using-ai-to-give-teachers-their-time-back
[Product Hunt, 2025] AI Grading Assistant for College Professors - GradeWiz | https://www.producthunt.com/products/gradewiz
[HolonIQ, 2023] Global Education Outlook | https://www.holoniq.com/notes/global-education-market-to-reach-10t-by-2030
[Global Market Insights, 2024] Online Homework & Assessment Platform Market Size | https://www.gminsights.com/industry-analysis/online-homework-assessment-platform-market
[Chronicle of Higher Education, 2024] The Grading Grind | https://www.chronicle.com/article/the-grading-grind
[Inside Higher Ed, 2024] The State of Academic Technology | https://www.insidehighered.com/news/tech-innovation/artificial-intelligence/2024/01/16/ai-academic-integrity-software-market
[Reuters, 2020] Instructure to be taken private in $2 billion deal | https://www.reuters.com/article/instructure-m-a-thoma-bravo/instructure-to-be-taken-private-in-2-billion-deal-idUSKBN28I2T5/
Articles about GradeWiz
- GradeWiz's AI Graded 30,000 College Papers Before It Left Y Combinator — Two Cornell undergraduates are selling a 97% accurate grading assistant to professors, betting they can scale a university pilot wedge into a real business.