I ship products end‑to‑end, and the agent systems that run them.
I'm Griffin Terry. I co-founded next4, a college-matching app on the App Store built over 6,271 institutions of federal data, and I'm the solo founder of Hirely. I do the whole arc: find the problem, model the data, build the product, then design the AI systems that distribute it.
An early version of our ranking treated "academically strong" and "socially active" as opposite ends of one scale — which quietly assumed a school can't be both. Splitting it into two measured indexes showed the real relationship:
correlation −1 (assumed)→+0.42 (measured)
Things I can show you.
Every figure below is traceable to a public source or a shipped artifact. None of it is rounded up.
| Figure | What it is | Where it comes from |
|---|---|---|
| 6,271 | Institutions in the knowledge base I sourced and modeled for next4 | IPEDS · College Scorecard · CDS · Carnegie · BEA |
| 145,000+ | Individual facts behind it, each carrying its own provenance | next4 platform, migration-verified |
| 1,226 | Reddit comments coded and adversarially verified before we wrote the product's code | 39 threads, 4-coder fan-out |
| 84 | Parents and students surveyed on demand, price and AI posture | Fielded Aug 2026 · n=69 / n=15 |
| 14 | Posts a week across six channels, produced by a marketing system I designed | next4 social system, running now |
Three things I built.
next4
Co-founder · App StoreThe college search runs backwards: a student checks a box on a test, their data goes on a list colleges buy, and 400 brochures arrive from schools that bought their demographic. next4 runs it forwards. A student sorts 86 preference cards; a deterministic engine ranks real schools against federal data and shows which of their own answers put each one there. No language model ever touches a school statistic, so it can't invent one.
Read the case study →
Hirely
Solo founderA tool that tailors your real résumé to a specific job without inventing experience. I architected and deployed the full-stack MVP on Supabase, Stripe and Vercel. In early beta, every waitlist signup since launch has used the Stand-Out Scan.
Read the case study →CoPilot
Growth & financeA two-driver rideshare that gets you and your car home. I owned the business model, pricing and five-year plan, pivoted the target segment after 100+ consumer interviews, and tied for first of twelve teams.
Read the case study →Getting an agent to produce something is easy. The work is what checks it.
I build with AI agents constantly. Almost everything worth showing came from the structure around them, not the generation itself.
I run marketing as an organization, not a prompt.
The verification step has to be able to kill the work.
Research earns the right to cut a feature.
Agents are confidently wrong, so review is a separate job.
Where I've built and led.
I'm a cum laude CU Boulder graduate — Political Science & Economics, with a Business minor — who builds products end to end. At next4 I cover product, the data layer and marketing. At Hirely, as solo founder, I built and deployed the whole thing myself.
What I care about most is honesty in the product. Hirely won't invent experience you don't have. next4 won't show an admission percentage it can't defend, and won't write a student's essay. The same rule governs the marketing: twelve non-negotiables sit between a draft and a channel, and they have stopped finished work from going out.
I gravitate toward the messy middle, where a vague problem has to turn into something people actually need — coding 1,226 Reddit comments before writing a line of next4, pivoting CoPilot's whole target segment after 100+ interviews. I'm open to roles across my areas of experience.
Co-founder, next4
Boulder, CO
Co-founded a college-matching iOS app, shipped to the App Store in September 2026. Sourced and modeled a 6,271-institution knowledge base from five federal datasets, drove the deterministic match engine and Expo app through Claude Code with spec-first, test-driven workflows, and ran demand validation across 1,226 coded Reddit comments and an 84-person survey before committing to the build. Designed and operate the agent-run marketing system behind distribution.
Founder & CEO, Hirely
Boulder, CO
Architected and deployed a full-stack MVP on Supabase, Stripe and Vercel using Claude Code. Drove 2,000+ impressions, 100+ profile visits and 60+ site visits in the first 14 days by automating LinkedIn, Instagram and TikTok distribution with AI agents. Pitched product vision and architecture to investors, users and peers.
AI Trainer, Handshake
Remote
Train and evaluate AI models, applying hands-on agentic-AI and product experience to improve model quality.
President, Up&Up Festival
Boulder, CO · Ambassador → Lead → President
Sold out a 500-person venue leading a cross-functional team of 16 across marketing, sales and social, running 10 tabling events and outreach to 5,000+ followers. Secured $1,000+ in sponsorships and built partnerships hosting events of 250+ attendees.
Director of Growth & Finance, CoPilot
Boulder, CO
Tied for first of 12 teams pitching to 50 peers and community members. Validated demand, pivoted the target segment after 100+ consumer interviews, and modeled five-year profitability.
Director of Event Management, Phi Kappa Psi
Seattle, WA
Directed a team of 80 to execute 50 events, six with 1,000+ attendees. Secured $6,000 in Red Bull sponsorship and partnered with the IFC and Washington State Liquor Board on security and venue logistics.
Education
CU Boulder · Universidad Carlos III de Madrid · University of Washington
B.A. Political Science & Economics, Business minor, CU Boulder — cum laude, 3.75, May 2026. International economics at Universidad Carlos III de Madrid, spring 2025. Dean's List: Winter '23, Fall '23, Spring '24, Fall '24, Spring '26.
If any of this is the kind of work you need done, I'd like to talk.
Open to roles across my areas of experience. Happy to walk through any decision on this site, including the ones that turned out wrong.