Why Colosseum Is the Best Hackathon — And Why That's Really a Story About Solana

A mentor named Simon leaned over our table in a Toronto co-working space, looked at the game running on our Seeker phone, and smiled. “This is really cool,” he said. Then: “I’ve got some bad news, and I don’t want to discourage you.”

We were two longtime Ethereum developers, deep in our first weeks inside the Solana world, and we’d spent a month building a game played entirely by AI agents. We’d come north straight from the EasyA hackathon at Consensus in Miami — Solana Mobile had sponsored it, so we were carrying Seeker phones — and walked into the Superteam office for the final days of the Colosseum hackathon, where the desks were open to anyone. We were proud of what we’d made. You could probably see it on us.

“Colosseum isn’t really interested in gaming,” Simon said. And before that could land as a verdict, he asked the question that took our whole project apart and rebuilt it: “Is there a specific problem you’re solving?”

We’ll come back to that question. It matters more than the answer — and it’s the reason we now think Colosseum is the best hackathon in crypto. But to explain why, we have to start with what most people get wrong about hackathons.

Most hackathons end. The best one doesn’t.

Ask anyone what makes a hackathon good and they’ll reach for the obvious levers: prize pool, swag, the energy in the room. Those describe a good weekend. But the thing that actually changes a builder’s life isn’t the weekend — it’s the Monday after, when the demo’s over, the judges have gone home, and the project that felt inevitable at 3am has to survive contact with the rest of your life.

Most hackathons are event-shaped. You win, you get a check and a handshake, and then you go figure out the other 99% of building a company alone. The hackathon was the destination.

Colosseum is company-shaped. The hackathon is just the qualifier.

In one line: Colosseum is the company that runs Solana’s official global hackathons, an 8-week startup accelerator, and a $60M venture fund — a single pipeline that turns a hackathon project into a funded company.

The fast facts (some self-reported by Colosseum — noted): Colosseum runs Solana’s global hackathons and an accelerator and a $60M venture fund. Every accelerator company gets a $250K standard check — a SAFE with a token warrant. 44 companies have been funded across its first four cohorts — at roughly a 0.67% acceptance rate. More than 80,000 builders have come through the hackathons, and by the organizers’ own account, about half had never built on a blockchain before.

Colosseum didn’t come from nowhere

Here’s the first turn of the story. Colosseum isn’t a startup that decided to run a contest. It’s the institutionalization of the single most productive developer-onboarding engine in crypto.

From 2020, a Solana Foundation growth lead named Matty Taylor built the Foundation’s hackathon program from scratch — eight straight online hackathons that pulled in 60,000+ builders and seeded companies like Tensor, Squads, StepN, and Jito. In January 2024, the Foundation did something telling: it spun the program out into an independent company — Colosseum, which Taylor co-founded with Clay Robbins and Nate Levine — with its own accelerator and fund, explicitly modeled as “the YC for Solana.” (Its seasonal hackathons now run under names like Renaissance, Breakout, and Cypherpunk.)

Taylor’s pitch for why a hackathon beats a pitch deck is the cleanest articulation of the whole thesis:

“Winning a five-week hackathon is a much better indicator of potential founder and founding team talent than a traditional written application.” — Matty Taylor, co-founder, Colosseum

So Colosseum built the rule directly into the machine: the only way into the accelerator is to win a hackathon. No warm intro. No pedigree. No cold email that gets you a meeting. You ship, in the open, against the world — and the work is the application.

The conveyor belt

Once you see it, you can’t unsee it. The Solana ecosystem isn’t a loose constellation of organizations — it’s a deliberately engineered conveyor belt that moves a person from “curious developer” to “funded founder,” and every handoff is documented:

Superteam            → local on-ramp: bounties, grants, build stations, fellowships
Colosseum hackathon  → global discovery (win = invite-only)
Colosseum accelerator→ $250K pre-seed, 8 weeks, demo day
Superteam Black      → post-raise scaling, perks, warm VC intros
Solana Ventures + VC → the capital layer

This is not metaphorical. In one Colosseum cohort, 8 of 13 teams came from the Superteam community. Superteam Black is literally defined as the rung above the accelerator, for teams that have already raised. Build Stations — the IRL co-working hubs — are timed to land exactly during Colosseum hackathon windows.

Which is how we ended up in that Toronto room. We didn’t know it at the time, but we were standing inside the on-ramp.

When Simon asked “what’s the problem?”, we did what builders do: we described what we’d built. “That’s a solution,” he said. “What’s the problem?” We tried again. “That’s a solution too.” Again, and again — gently, relentlessly — until the floor stopped feeling like a floor and started feeling like a door. Then the rest of the Superteam crew — Ben, Julian, and the others — kept pushing alongside him.

And somewhere in there we heard ourselves say the thing we’d built without ever naming it: our agents needed private, encrypted memory they could own and license — data you’re not afraid of leaking. We’d built it as plumbing for a game. Said out loud as a problem, it changed shape: as the agent economy arrives, every autonomous agent will need exactly that. In the last 48 hours, with the whole Superteam team helping us tighten the pitch, we tore the project down to that core and rebuilt it. That’s how Vellum Network was born — and we went on to place 2nd in the Colosseum Canada track.

That’s our story. But a good story isn’t evidence. Here’s the evidence.

What comes out the other end

A few of the companies this machine has produced — across the Foundation-era hackathons and the Colosseum era. Each gauge below is one project’s own standout number, not a ranking. Different companies, different yardsticks, all of them started in a hackathon.

StepN
4th place, $15K hackathon entry

Jito
liquid staking + MEV

Squads
multisig for 300+ teams

Tensor
bootstrapped from ~$65K in prizes

ORE
Colosseum Renaissance champion

Marinade
launched on an ~$80K grant

DeCharge
EV-charging DePIN · Cohort 1

The breadth is the point: DeFi, infrastructure, consumer, DePIN, payments. And the Canadian thread runs right through it — TapeDrive, a decentralized-storage team, won the Breakout Grand Championship and made the accelerator. These aren’t decorations. They’re proof that the weekend doesn’t decay into a Monday-morning hangover. It compounds.

A different kind of hackathon — not a better one

To be clear, “company-shaped” isn’t a knock on event-shaped hackathons; the best of them are extraordinary. The clearest example is ETHGlobal, the gold standard of the Ethereum world — its in-person events are some of the best hackathons anywhere, and we say that as longtime Ethereum builders ourselves. Colosseum and ETHGlobal simply aren’t built for the same thing:

Different ecosystems, different formats, different strengths — each the best at what it does. If you’re building on Ethereum, ETHGlobal is where you want to be; if you’re building on Solana, it’s Colosseum. This piece is about the Solana side of that map — and what makes its on-ramp unusually powerful.

The reveal: it was about Solana all along

So Colosseum is a great hackathon with a great pipeline. But pipelines need a reason to exist — and this is where the story turns one more time.

Colosseum, Superteam, the accelerator, the $60M fund, Solana Ventures — these aren’t independent organizations that happen to cooperate. They’re the deliberate expression of a single, almost paradoxical strategy: the Solana Foundation is trying to make itself unnecessary. At Breakpoint 2024, its executive director literally staged a formal debate arguing that the Foundation should be dissolved. Its stated 10-year goal is its own redundancy. So it externalizes everything that matters: it spun out its core engineering (into Anza), its builder pipeline (into Colosseum), and its community (into Superteam) — each one an institution built to replace the institution.

That philosophy was forged in fire. When FTX collapsed in November 2022 and SOL fell 97%, the obituaries were already written. But the network never stopped producing blocks, and nobody left.

“A lot of people left Solana for dead… But what never really changed was no one left and everyone kept building.” — Lily Liu, President, Solana Foundation

“The fact that the network went through that shitshow and just continued running and people continued building it — it’s the most decentralized thing possible.” — Anatoly Yakovenko, co-founder

The mechanisms that carried Solana through the trough — the hackathons, the community, the builder pipeline — are exactly what Colosseum later institutionalized. And the reason all of it is even possible is the technology underneath. Anatoly Yakovenko’s insight, scribbled out at 4am on “two coffees and a beer,” was that time itself could be a data structure — Proof of History — which is why a Solana transaction costs about $0.00025 instead of dollars. That sub-cent, sub-second economics is what makes a new class of software viable: real-time on-chain games, micro-subscriptions, and software that pays software.

Which is the frontier Colosseum is now running straight at. In February 2026 it ran a hackathon where the contestants were AI agents — humans were forbidden from writing code; the agents registered, formed teams, and shipped on their own. It’s not a gimmick. It’s the same machine, pointed at the next economy: the one where agents hold budgets, pay per call, and need data they own. (That, as it happens, is exactly the problem Simon grilled out of us.)

So who’s the protagonist of this story? Not Colosseum. Not even the ecosystem. It’s Solana — a chain whose institutions are engineered to turn an anonymous developer anywhere in the world into a funded founder, and then get out of the way.

So — why is Colosseum the best hackathon?

Because of everything it’s the front door to.

The prize pool isn’t why. The best hackathon is the one that doesn’t end — and Colosseum doesn’t end because the entire ecosystem behind it is built so it doesn’t have to. Win, and you don’t get a handshake; you get a $250K check, an accelerator, a scaling program, a venture layer, and a community whose mentors will sit at your table in Toronto and ask the one question that rebuilds your company.

We walked into that room as Ethereum developers, tourists in someone else’s ecosystem, holding a game. We walked out with a real company, a Canada-track finish, and a different plan for our lives. We’re not visiting Solana anymore. We’re building here full-time.

That’s the honest measure of a hackathon: not what you win, but who you become. By that measure, Colosseum is the best one going.

A note on honesty: we’ve stuck to numbers with primary-source backing. A few widely-quoted Colosseum figures (like “80% of venture-backed Solana startups began in hackathons”) are the organization’s own, unaudited — we’ve flagged those as claims. And Colosseum is Solana-specific and genuinely selective: winning a hackathon is a qualifier, not a guarantee. That’s the right framing for the real claim — it’s the best hackathon for builders who want to become founders.

FAQ

Written by Liam C. and Claude-do for Superteam Canada. Figures reflect public sources as of mid-2026; self-reported numbers are attributed as such.