
To hire a head of engineering for an 8 to 30 person startup, expect a 90 to 150 day search, a $250k to $400k US base salary plus 1 to 2 percent equity, and a hiring bar built around four things: written communication, hiring chops, calm in crisis, and willingness to still write code occasionally. This is the "first VP Eng" decision, usually made between Series A and Series B, and it sets your engineering culture for the next three years.
Most founders hire too early or too late, almost never on time.
Too early looks like a 4-person team where the CTO hires a head of eng to "free me up for product." The new hire spends six weeks looking for things to manage and ends up writing process docs no one reads. Too late looks like a 25-person team where the CTO is doing 1:1s 30 hours a week, three engineers just quit, and the roadmap is two quarters behind.
The right window is 8 to 30 engineers, almost always within 12 months of your Series A. At 8 people you have enough scope to justify a dedicated leader (two squads, hiring pressure, an on-call rotation that needs structure). At 30 people you have probably already missed the window and you are now hiring a VP Eng to clean up debt rather than build forward.
If you have a strong CTO who wants to stay technical, the first leadership hire is usually a head of engineering, not a VP. The head of eng owns people, process, and roadmap. The CTO owns architecture and external technical credibility. If you have a non-technical CEO and a part-time technical advisor, you might want a fractional CTO before a full-time hire, at least until you have product-market fit signal.
These four titles get conflated constantly. They are not interchangeable, and getting the scope right at the offer stage saves you a painful 90-day correction later.
| Role | Team size fit | Comp (US, 2026) | Time on code | Primary scope |
|---|---|---|---|---|
| Head of engineering | 8 to 30 engineers | $250k to $400k base + 1 to 2% equity | 10 to 30% | People, hiring, roadmap execution, eng process |
| VP Engineering | 30 to 150 engineers | $350k to $550k base + 0.5 to 1.5% equity | 0 to 10% | Org design, multi-team coordination, exec leadership |
| CTO | Founding to 50+ | $200k to $400k base + 2 to 10% equity (founder-level) | Varies wildly | Technical vision, architecture, external credibility, board |
| Principal engineer | Any | $300k to $500k base + 0.25 to 0.75% equity | 60 to 80% | Deep technical scope, no direct reports, IC ladder |
A head of engineering is closer to a "player coach." They run sprint planning, do 1:1s with 6 to 12 engineers, own quarterly OKRs, and still open pull requests in week one to earn credibility with the team. A VP Eng at a 100-person company has not touched the codebase in two years and that is correct for the role.
The compensation gap matters. If you pay VP Eng money for a head of eng scope you will attract someone who wants to manage managers, not write code. They will be miserable in a 12-person org and leave in 14 months.
We have helped founders run dozens of these searches. The signal that actually predicts success at this stage clusters around four traits. Everything else is noise.
A head of engineering at a 15-person startup writes more than they speak. Sprint retros, design doc comments, hiring rubric updates, board prep, async standups, post-incident reviews, the weekly engineering update to the rest of the company. If they cannot write a clean 400-word doc that a non-engineer can follow, they will struggle.
Ask for two artifacts in the process: a real engineering update they sent at a previous company (redacted is fine), and a post-incident review they wrote. Read both carefully. If the writing is bullet salad with no narrative, that is your answer.
At your stage, the head of eng will personally close 6 to 15 engineers over the next 18 months. That requires actual recruiting work: sourcing, calibrated interviewing, writing offers that get signed, and building a hiring rubric that scales beyond their own taste.
Test this with a working session. Hand them a real job spec you are about to post (say, a senior backend engineer) and ask them to redesign your interview loop in 45 minutes. Watch how they think about signal versus noise, how they handle the trade-off between speed and quality, and whether they push back on your stated requirements when they should. Our Series A engineering hiring playbook covers the loop design in more depth.
Production goes down. A founding engineer rage-quits the day before a board meeting. A customer threatens to churn over a security incident. These happen at 15-person startups every quarter, sometimes every month.
Ask three behavioral questions, each anchored to a specific past incident:
You are not looking for textbook answers. You are looking for someone who sounds neither defensive nor performatively calm. The good ones describe the chaos honestly and then explain what they actually did, including the parts that did not work.
This is the trait that separates a head of eng who thrives at 15 people from one who washes out. The job is not pure management at this stage. They need to pair with an engineer on a thorny bug, prototype a spike when a feature is blocked, review architecture PRs with real opinions, and ship the first version of internal tools when no one else has the bandwidth.
Ask: "When was the last time you opened a pull request that shipped to production? What was it?" If the answer is "two years ago," they have moved past the head-of-eng phase of their career. That is a VP Eng candidate, not your candidate.
Every engineer on Cadence is AI-native by default, vetted on Cursor, Claude Code, and Copilot fluency before they unlock bookings. A modern head of eng should be comfortable using those tools themselves and should have an opinion about how the team uses them. If they treat AI coding tools as "what the juniors use," that is a yellow flag in 2026.
Most heads of eng are not actively looking. The good ones are 18 to 30 months into a role they like, with vested equity. You have to go get them.
| Channel | Pros | Cons | Best for |
|---|---|---|---|
| LinkedIn direct outreach | Reaches passive candidates; you control the pitch | High volume of cold messages; 5 to 10% response rate is normal | Most searches |
| Founder network referrals | Pre-vetted by someone you trust; faster close | Limited pool; can create echo chamber | First 20 candidates |
| Executive search firms | Outsourced sourcing; structured process | $50k to $120k fee; 4 to 6 month timelines | If you have no network and no time |
| Y Combinator Work at a Startup | Pre-filtered for startup tolerance | Skews junior; few true VP-level candidates | Series A backed companies |
| Pallet / talent collectives | Curated lists from operators you respect | Small inventory; hit or miss timing | Opportunistic |
| Cadence (booking, not hiring) | Book a Lead tier engineer at $2,000/week for fractional CTO scope; 48-hour trial | Not a full-time placement; better for interim than permanent | Bridging the search; trying a working relationship before offering full-time |
The most common mistake here is going straight to a retained search firm. That makes sense once you have run 20 sourced conversations yourself and know what you actually want. Before that, the search firm will produce a generic spec and a generic candidate.
If you are 90 days into the search with no offer extended, the answer is rarely "hire a recruiter." The answer is usually "your spec is wrong." Recalibrate on team size, on whether you want a player-coach versus a pure people leader, and on what trade-offs you are willing to make on tech stack experience.
A head of eng interview loop should take 4 to 6 hours of candidate time across 2 to 3 weeks. Anything shorter and you do not have enough signal. Anything longer and your best candidates drop out.
If a candidate is uncomfortable with the working session, that is real data. The role itself is mostly working sessions.
US comp for a head of engineering at a Series A or early Series B startup in 2026 looks like this:
If you are based in lower-cost markets (Austin, Denver, Toronto), trim 10 to 15 percent off the base band. If you are hiring remote-first from outside the US, the global market for this role is $180k to $280k base for equivalent quality, with equity bands roughly the same.
A note on the search itself: budget $30k to $60k in your own time (founder and existing leadership) over the 90 to 150 day window. If you use a retained search firm, add $80k to $120k in fees. The all-in cost of a bad hire here is around $400k to $600k once you count the search, severance, the 18-month productivity drag, and the engineers who quit on the way out. Hiring slow is almost always cheaper than hiring fast.
Most founders treat the head of eng decision as binary: hire full-time, or do without. There is a third option.
If you are 60 to 120 days into the search and have not found the right person, book a Lead tier engineer (typically $2,000/week on Cadence) to act as fractional CTO or interim head of eng for 8 to 16 weeks. You get someone running sprint planning, mentoring your team, and helping you write a better spec for the full-time hire. You also get a working sample of what a strong technical leader looks like in your environment.
This is not a replacement for the full-time hire. It is a bridge. The fractional engineer can sometimes convert to full-time if the fit is right, but more often they hand off to a permanent hire after 90 to 120 days, and you have a much better interview loop because of it.
If you want to see how this works, Cadence's founder onboarding walks through booking a Lead tier engineer in two minutes with a 48-hour free trial. No notice period, replace any week, weekly billing. For a role this senior, the 48 hours is genuinely enough to know whether they can hold a room of engineers.
90 to 150 days from kicking off the search to a signed offer, assuming you are sourcing actively and running a 4 to 6 hour interview loop. Searches that drag past 180 days almost always have a spec problem, not a sourcing problem.
If your engineering team is under 30 people, head of engineering. If you are scaling from 30 to 100+ in the next 18 months, VP Eng. The titles map to scope, not seniority; hiring the wrong scope creates a 12 to 18 month correction.
At an 8 to 30 person startup, yes. Expect 10 to 30 percent of their time on code in the first six months, dropping toward 10 percent as the team grows. A head of eng who refuses to write code at this stage is a VP Eng in the wrong scope.
1.0 to 2.0 percent for a post-Series A hire, four-year vest with a one-year cliff. Above 2.0 percent only if you are pre-Series A or replacing a co-founder. Below 1.0 percent will lose you the best candidates to your competitors.
A head of engineering manages people, owns the roadmap, and reports to the CEO. A principal engineer is a senior IC with no direct reports, owns deep technical scope, and reports to the head of eng or CTO. You usually want both, in that order, once you cross 15 engineers.
Runs the talent acquisition manager bench at withRemote. Writes on interviewer calibration, offer mechanics, and TA team operations.