For each of these 10, we start with the impact the project must create — then work backward to exactly what it needs to be. The scores below are from all 5 admissions experts.
Each project below was scored across 5 AO lenses: Elite Private, Large Public, CS Specialty, HS Counselor, and Diversity/First-Gen. We then applied the impact-first filter: Who benefits? How measurably? What must the project actually be to create that impact?
Who is helped, how, and why does it matter beyond the student's résumé?
What must the project technically and tactically be to generate that impact?
Can a HS student realistically build this in 3–18 months?
Engineering? Leadership? Research? Community building? Advocacy?
40 kids from underrepresented backgrounds complete a 2-week CS camp and 80% report increased interest in pursuing CS. The camp becomes an annual school program run by future students — sustainability after the founder graduates.
For this impact to be real: the camp must be free to attend (removes access barrier). The curriculum must be original — not just Scratch tutorials. The student must train and manage 10+ HS volunteers. The school must formally adopt it. Each camper's outcome must be tracked with parental consent. The 80% "increased interest" must be measured via pre/post survey.
Elite Private: "Interesting but needs to show engineering depth alongside the management. What was the student's technical role beyond organizing?"
Large Public: "This is exactly what we look for — a student who identified a gap and built a lasting solution in their own community."
CS Specialty: "Curriculum design is genuinely technical work. But the camp needs a substantive technical project — not just block coding."
Counselor: "The retention data — 80% increased interest — is the key proof point. The harder question: who came back for more?"
Diversity: "My highest-scored project on the board. The sustainability — school adopts it annually — is the difference between a project and an institution."
An official Girls Who Code chapter affiliated with the national organization, growing from 0 to 30+ active members. 50+ girls complete the full curriculum. 10 members enroll in AP Computer Science the following year — a measurable pipeline change documented by the school.
Must be an officially affiliated GWC chapter (not "inspired by"). Must track enrollment, completion, and AP CS continuation with counselor sign-off. The student's role as founder and primary instructor must be documented. The curriculum must be original work — not just the GWC standard curriculum. And the AP CS enrollment number must be real.
Elite Private: "Strong. 'Founded' is one of the most powerful words in an application. The 10 AP CS enrollments are the proof point."
Large Public: "This aligns directly with our institutional mission around access. Official affiliation is critical — it shows the student navigated an external organization."
CS Specialty: "The concern: is this leadership or participation? I want to see what the student personally built vs. facilitated. The curriculum they designed is the technical evidence."
Counselor: "One of the most reliable project types for essays — the founding story writes itself. But the 10 AP CS students need to be named with permission."
Diversity: "Solid but slightly lower than the camp — GWC chapters exist in many schools. The differentiator is whether this student made it something it wouldn't have been without them."
200+ lost items reunited with their owners in one school year. The AI image-matching system handles 50+ queries per day during peak periods. The full school district adopts it — 3+ schools — reducing lost property claims by 80%.
Must include computer vision (CLIP or similar) for image similarity matching, not just text search. Must have a real district adoption story — the student had to convince the district IT department. Must have a quantified match rate. Privacy must be handled: FERPA compliance, no facial recognition, data retention policy published.
Elite Private: "Decent but not exceptional. The technical claim needs to be stronger — what specifically did the AI do that a database couldn't?"
Large Public: "This is exactly the kind of civic technology project we celebrate. District adoption is the proof of institutional credibility."
CS Specialty: "CLIP-based matching is the right technical choice. I want to know: what was the precision/recall? What were the failure modes?"
Counselor: "The story is straightforward but solid. The 200-item number needs a documentation trail — before/after comparison with prior year's lost property."
Diversity: "Strong. The district partnership required navigating institutional bureaucracy — that skill is often overlooked but deeply valuable."
A management platform for school clubs — events, attendance, elections, communication — built for and adopted by the student's own club, then expanded to 15 clubs covering 500+ students across the school. The student persuaded 14 other clubs to switch, documenting the adoption process.
Core features: event scheduling, attendance tracking, officer elections, announcements. Must include evidence of adoption beyond the student's own club. The "sell" — convincing 14 club presidents — is the leadership story. The platform must be actively used, not just installed. Analytics showing real usage data required.
Elite Private: "The engineering is somewhat basic — database CRUD apps are common. The differentiator is the 500-user adoption story."
Large Public: "Perfect institutional fit score from me. This student built capacity in their school — that's exactly what we look for in our graduates."
CS Specialty: "Lowest technical score. This is management software. I want to see algorithmic complexity or systems design, not just CRUD operations."
Counselor: "Good story, decent numbers. The key question in the essay will be: what was hard? If the student can't answer that, the project loses weight."
Diversity: "Neutral on this one — it's a well-resourced school story. Not bad, just not particularly differentiated."
100+ high school tutors onboarded and verified. 300+ younger students matched and tutored. $5,000+ in combined tutor earnings flowing through the platform. Real payments processed — even token amounts — proving genuine economic value.
Must include: tutor verification (teacher recommendation required), scheduling system, messaging, and payment mechanism. The student must demonstrably recruit 20+ tutors — building the tech is not enough. Revenue must be real (Stripe or equivalent). Retention rate matters: are tutors and students coming back?
Elite Private: "Revenue is interesting but doesn't move the needle much at top schools. The recruitment story is what matters."
Large Public: "Community impact score is very high. This student identified a local need — affordable tutoring — and built a real solution."
CS Specialty: "A marketplace is technically interesting but not particularly deep. The hard part is the matching algorithm and payment integration."
Counselor: "The $5,000 number is solid proof. But I want to know: what happened to the students who got tutored? Did grades improve?"
Diversity: "One of the best community stories on the board. Free or reduced-cost tutoring for younger students is precisely the kind of local impact that matters."
3+ merged pull requests in real, actively-used open source projects. Each PR touches meaningful code — not just documentation fixes. The code runs in production in applications used by millions of developers. Release notes or changelog acknowledges the contribution.
Must show: PR links, the actual code changes, review feedback received and responded to, and the release/version it shipped in. "I fixed a bug" is interesting; "my fix shipped in Python 3.12 and affected every developer using that library" is remarkable. Target projects: Rust, Python, Node.js, React, Vue.
Elite Private: "3 PRs in major projects is genuinely impressive. The review conversation with maintainers is the intellectual evidence."
Large Public: "Solid but doesn't scream 'community impact.' For our mission, the CS camp or tutoring marketplace would score higher."
CS Specialty: "My highest technical score. This proves the student can write production-quality code that professionals trust. The code review history is an automatic credibility boost."
Counselor: "The story needs work — the essay can't just link to PRs. What did the student learn? How did they respond to criticism? That's the human story."
Diversity: "This is the great equalizer — you need nothing but a laptop and internet. That story matters in my evaluation."
300+ GitHub stars from an open source homelab dashboard project. Mentioned on the Homelab subreddit (not just posted, but organically discussed). Referenced in 2+ YouTube homelab setup videos. The project is used by homelab enthusiasts worldwide — not just the student's own setup.
Must be genuinely open sourced (not just a GitHub repo that exists). The 300-star threshold forces real quality and real promotion. Must have documentation, installation instructions, and a clear differentiation from existing homelab dashboard projects. Stars must come from genuine users, not bot accounts.
Elite Private: "Underrated project. The combination of systems thinking and aesthetic sensibility is rare. 300 stars is real community validation."
Large Public: "Decent — the homelab community is real but niche. Not the strongest community impact story."
CS Specialty: "Systems administration + beautiful UI is an underrated combination. The student has to understand networking, services, and APIs simultaneously."
Counselor: "The essay angle is solid — 'I wanted a dashboard that looked the way my server felt.' That's memorable. The 300-star number must be real."
Diversity: "Homelab requires equipment. I deduct slightly for the resource requirement — not everyone can run a server. But it's still accessible relative to most hardware projects."
5 computer science teachers using the system for grading. 2,000+ assignments auto-graded with test suites. Integration with GitHub Classroom — so teachers don't change their existing workflow. The student got 5 teachers to change how they grade, which is a fundamentally different skill than building a tool.
Must integrate with GitHub Classroom API (real LMS integration). Must handle Python, Java, and JavaScript. Must run in Docker sandbox. Must detect plagiarism patterns. Must provide actionable feedback, not just "wrong." Test suites must be reviewed by teachers. The 5-teacher adoption number is the proof point.
Elite Private: "Engineering is solid but somewhat narrow. The 2,000-graded-assignments number is interesting but not exceptional."
Large Public: "Good institutional fit. This student made teachers' lives easier — that's a service orientation we value."
CS Specialty: "One of the most technically demanding projects on the board. CI/CD pipelines, Docker sandboxing, LMS integration — this student knows production engineering."
Counselor: "The adoption story is what matters here — how did the student convince 5 teachers to try something new? That's the leadership evidence."
Diversity: "Neutral. No particular resource barrier or community impact dimension that differentiates it."
School-board approved and counselor-endorsed AI check-in system. 100+ students use it weekly. 12 real crisis situations detected and escalated to professional counselors with proper handoff. Zero data breaches. The student navigated safety-critical software engineering with institutional oversight.
Must use a validated screening instrument (PHQ-4 or equivalent). Must have counselor escalation with explicit protocol. Must be approved by school board and legal counsel before launch. Must have no PII stored. The 12 crisis detections must be documented with counselor sign-off. The school-board approval letter is the credibility document.
Elite Private: "The institutional approval is impressive — navigating bureaucracy at this level shows maturity. But the AI piece needs to be more technically rigorous."
Large Public: "Mental health is a top priority at our institution. This student identified a real need and built a solution that passed legal review."
CS Specialty: "Safety-critical systems design is the technical challenge here. PHQ-4 integration and escalation logic require understanding both the clinical domain and software architecture."
Counselor: "This is the kind of project I wish more students attempted. The essay story is powerful — why did they care about this? What surprised them? What would they do differently?"
Diversity: "Mental health support in schools is disproportionately absent in underserved communities. The community impact is real even without an explicit underserved focus."
500+ students used the tool for real college essays. Multiple documented acceptances at named top-choice schools (with permission). The tool's feedback quality was validated in a blind comparison study against professional college counselors — with measurable correlation between tool-assisted essays and admission outcomes.
Must be trained on or prompted for AO perspective specifically — not generic text feedback. Must have a formal validation study: 20 essays rated by both AI and a counselor, with documented correlation. The 500-user threshold must be real (analytics). Testimonials must include school names (with permission). The validation study is the key credibility document.
Elite Private: "Strategically clever — the tool evaluates exactly what we do. But the validation study is essential. Without it, this is just another AI tool."
Large Public: "Good community impact story. College application stress is universal — this student built something genuinely useful for their peers."
CS Specialty: "The NLP is interesting but not deeply technical. The validation methodology is the more rigorous piece."
Counselor: "Highest counselor score on the board. The best version of this project: the student used their own tool on their own essays and got into their dream school. That's an unforgettable essay."
Diversity: "The 500-student number includes students from all backgrounds — that's good. But the validation study is what matters for equity: does this tool work equally well for first-gen applicants?"