--- title: Do I Need a CTO Yet? url: https://varstatt.com/jurij/p/do-i-need-a-cto-yet author: Jurij Tokarski date: 2026-08-04 updated: 2026-10-25 description: A non-technical founder is told to find a CTO and give up equity. Early on you need the calls and the build, not the title; who makes them decides the equity. section: Blog (https://varstatt.com/jurij/archive) tags: Q&A (https://varstatt.com/jurij/c/questions) --- I'm a non-technical founder with a validated idea and a little funding. Everyone says I need a technical co-founder or a CTO before I build anything. Do I, and do I have to give up equity to get one? You need the work, not the title. Someone has to make the technical calls and build the first version, starting now. Who does it decides whether equity changes hands. A technical co-founder takes shares, and a first hire a salary, often with options. Someone paid by the week takes cash, and your cap table stays as it is. The title starts to matter later, once there are engineers to lead. ## The First CTO Job Is Building With nobody to manage, a CTO picks the stack, shapes the data and builds the first version. That's a founding engineer's job. Founding engineer and fractional CTO are two names for that one seat. It covers everything technical before there's a team, from the stack to [monitoring](/principles/diligence/monitor-day-one), in three jobs: - Decide: the stack, the architecture, build or buy, what comes first. - Build: working code, shipped every week. - Watch: what breaks, what can't be undone, what can wait. What's left out is the management half of the title: hiring, team management, board updates, investor decks. Until your first hire, that seat is the whole technical job. That leaves your own worry: if the job is this technical, can someone who isn't own it? ## You Own the Problem, Not the Code You don't need to be technical. You need to own the problem and make the decisions. Your part is the business side of every call: who it's for, what comes first, what a trade-off is worth. The technical side turns each choice into options, with the reasons in plain words. Then you decide. Written down, one of my calls reads like this. "Payments through Stripe Checkout, not a custom form: less to build, and no card data on your servers. Easy to swap later. Your call." If nobody in the company makes calls like that, then yes, you need a CTO, whatever the title. Each call goes into the repo in writing, so the [reasoning outlives the conversation](/principles/diligence/documentation). Your side starts before anyone is hired. Write down [what to send a developer first](/jurij/p/what-makes-a-good-project-brief): the problem, who has it, what done looks like. The free [project scope tool](/discovery/project-scope) turns that into user stories a developer can read. What's left is who makes the technical calls, and how they get paid. ## Equity Buys a Partner, Not Labor A co-founder, a first hire or a fractional CTO can make the same calls. What differs is how you pay and how long you're tied together. A co-founder is paid in shares, usually vesting over years. They own part of the company and a say in its direction. The deal is hard to undo when the match fails. A fractional CTO is paid in cash for the weeks it covers, owns nothing in the company and is easy to end. A co-founder brings one thing a paid seat can't: an upside tied to yours, through the bad quarters too. That's worth shares when the person is right. It's a costly way to buy the first build when they're not. Equity isn't a way to skip paying an engineer. Plenty of good engineers hold equity, and I'm not against it. What fails is equity offered as payment for the build, from a founder who wants free labor, not a co-founder. In 2025 I wrote about LinkedIn posts with [99 requirements for a technical co-founder](/jurij/p/when-optimization-culture-breaks). They promised equity for a free MVP. I answered one with the questions a real co-founder would ask, and the reply was "Thanks, not interested." That equity usually vanishes, because most of those projects die. The offer often works as a test of seriousness. A founder who means to do the work knows the idea needs paid labor too. Co-founding is something else. A co-founder brings business judgment as well as code and takes a stake in the real business. That's a partnership, not a payment. ## Test a Co-Founder on Real Work Finding the right person is the slow part. A match has to last years, so the best evidence is work done together, not a pitch. Start with people you've worked with. Test each one on a small piece of real work before anyone talks about shares. One of my own partnerships began as an advice ask to a university peer. A barter, a project and a build together came after. I've sat on a startup's founding team myself, for about two years. ## Bring One In Before the Code The time to bring one in is before the first line of code. The stack and the data model cost the most to undo. Both get decided in the first week, on purpose or by accident. In one app of my own, each user's feeds sat under that user in Firestore, where a path can't be moved. Sharing a feed between users would take a copy and a rewrite of all of it. So the redesign settled that path first. Later works too, when a first version has [no technical owner](/jurij/p/is-my-vibe-coded-app-ready-for-production). Your search and the build can run side by side: hire the work by the week and keep looking. The code and accounts stay yours, so a co-founder who joins later starts from a working product. Cash by the week raises the next question: how much, and for what. ## Compare What the Price Covers A fractional CTO's fee can mean advice only, or advice and the build. Compare that before the number. The classic setup is an advisor for a few hours a week plus a developer who builds. That's two bills, and you translate between them. The total depends on how many weeks the scope takes. The free [build cost tool](/discovery/build-cost) estimates them, so a little funding becomes a number of weeks. [What an MVP should cost](/jurij/p/how-much-does-it-really-cost-to-build-an-mvp) explains why quotes for one scope differ. Some founders hope to skip the fee and let AI do the job. ## AI Types Faster, the Calls Stay Human AI writes and reviews code. It doesn't decide what to build, or carry the cost when that's wrong. Reviews and typing were never the hard part. AI makes me 2 to 3 times faster at writing code, and the calls stay mine. Software is an iceberg. AI writes the visible part fluently. What's underneath keeps a product alive, and knowing where it's buried is the job. The seat is shifting all the same. Five years ago, this was CTO-level work in a small company. It will probably become [the normal expectation for a senior developer](/jurij/p/the-fourth-evolution-of-mvp). The judgment stays, and it [still decides what gets built](/jurij/p/domain-experts-still-need-tech-partners). ## What the Seat Costs Elsewhere A full-time CTO sets the top of the range. Kruze Consulting's payroll data covers more than 250 venture-backed startups in the United States. In 2024, a seed-stage CTO there earned $146,000 a year on average, and a Series A CTO $223,000. Upwork shows the other end. Of 81 CTO, founding-engineer and technical co-founder jobs open there in October 2026, 60 were posted by founders. Most offered hourly pay, at a median of about $34 an hour. So founders already buy the seat for cash. An hourly rate buys presence, though, and [hours are the wrong unit](/principles/philosophy/hours-wrong) for judgment. ## How I Sit in the Seat I sit in it as a [fractional CTO](/x/founders/fractional-cto). I make the calls and write the code, so nothing gets lost between an advisor and a developer. My price is one weekly rate for the decisions and the build, [listed on the homepage](/). No hourly billing, and you can [pause or cancel](/principles/partnership/exit-freedom) any week. When you hire a team, I stay as the senior engineer or hand over. To start, the [project form](/start?cta=cto-post) takes about a minute. ## What the Founder Should Do Now Get the calls made from the first week. Hire a full-time CTO when there are engineers to lead. Until then, a full-time salary also pays for management, with nobody to manage yet. With a little funding, it can take most of it. The switch to a full-time CTO is the expected arc, not a failure. One person covers a product until it outgrows them: three features at once, a team to hire, a board to report to. That's the limit the [solo model](/principles/philosophy/solo-model) names. The handover is cheap when the repo, accounts and keys [were yours from day one](/jurij/p/who-owns-the-code-my-developer-writes). Written decisions make it cheaper still: [access, docs, history](/jurij/p/isnt-a-solo-developer-a-bus-factor-of-one). So you don't have to give up equity to start. The calls need making this week, by someone you can stop paying any week.