How to Hire Your First Marketing Engineer
The scorecard, the JD you can steal, the comp band, and the first 90 days.
Profound broke the internet last week by naming the role that is quietly reshaping how marketing gets done. The Marketing Engineer. Nick Lafferty, first to officially hold it.
The title is new. The person is not. Every high-performing marketing team already has one, hiding inside other titles, doing the work without a name for it.
Here is one in the wild. At Webflow, I wanted to deepen acquisition in the freelancer community. Between days stacked with calls, I built a handful of agents that scraped freelancer and web dev subreddits, surfaced real-time pain points, shaped a narrative, shaped an offer, activated across channels, measured, refined. An executive with a toolbox, running an end-to-end cross-functional acquisition program in days, not quarters. Not a team. One person with the right tools and the range to use them. By the time I left Webflow, more of my team was operating this way than wasn’t.
The role is builder and artist in the same seat. The builder half writes the agent, picks the model, ships the dashboard, retires the vendor contract. The artist half decides what is worth building, what the narrative needs to say, and whether the output is any good. Neither half works alone. A marketing engineer who can only build becomes a backlog. One who can only imagine becomes a PM. The job is both, and the people who do it well are rarer than the title makes it sound.
This matters now because the functional split inside marketing is breaking down. Technical and creative. Strategist and operator. Analyst and storyteller. The lines are blurring because the best people stopped honoring them. They started shipping work that made the specialists look slow, and the org charts are catching up.
Early-stage companies have run this way forever out of necessity. The difference in 2026 is it is moving upmarket fast. The orgs winning already have this person inside them. The ones that don’t are feeling the gap.
In 2027, this is a standard role. In 2028, it is a team. The companies hiring the first one right now are the ones who will have the second and third already in seat by the time everyone else writes their first req.
Profound named the role. Nobody has written the definitive playbook for hiring one. This is that piece. The role, the req, the comp, and the first 90 days. For the VPs and marketing leaders making the hire. And the founders signing the offer.
Why the old archetypes don’t cover this
The growth marketer you hired in 2022 was a channel operator. Paid, lifecycle, SEO. Ran campaigns against a plan.
The marketing ops hire you made in 2023 was a systems operator. Shipping in Marketo, CRM hygiene, attribution, lead routing. Kept the pipes clean so the campaigns could run.
Both roles are still valuable. Neither one owns the loop.
The marketing engineer closes the loop alone. Identifies the problem. Builds the thing. Ships it. Measures it. Kills it or scales it. What used to be a cross-functional project with a PM, an engineer, an analyst, and a marketer is now one person’s week.
The teams that figure this out first run with a third of the headcount and ship four times the work. That is not a productivity gain. That is a different operating model.
That’s the math. That’s why you’re hiring.
What they actually do
The job is not “AI for marketing.” That is the vendor pitch. The work is more specific and harder to fake.
They deploy agents as production systems. Not Zaps. Not prototypes. Real systems with branching logic, model calls, data enrichment, human review, and measurable outputs. Research agents that pull competitor signal every morning. Enrichment agents that keep the CRM clean without a human touching it. Outbound agents that write like a human because a human wrote the pattern once. The marketing engineer owns reliability, unit cost, and output quality. When an agent breaks at 11pm on a Sunday, they are the one fixing it.
They build internal tools that retire vendor contracts. A lead scoring tool that beats the one you are about to renew. A content brief generator that pulls from your research library. A competitive intel dashboard that updates itself nightly. A first-party attribution view that reflects how your business actually works. Things that used to be a six-month engineering ask. They ship in days.
They own the measurement layer. They do not wait for analytics. They pull from the warehouse, the CRM, the product, and the model layer, and they build the view the team actually uses. They are the person who tells you, unprompted, that your attribution is lying and here is the query that proves it.
They work across every marketing function. PMM, content, demand gen, brand, paid, PR, lifecycle, localization. The agents and tools they build deploy wherever the leverage is highest that week. A launch analysis for PMM on Monday. A localization pipeline on Wednesday. A press monitoring agent on Friday. The breadth is the job.
They think in architecture, not tasks. The best ones see a single workflow and recognize the shape of the hundred others that run on the same structure. They build the first one right so the next ninety-nine compose instead of compound into debt. This is the cognitive attribute that separates a great marketing engineer from a fast one. It is also the hardest thing to assess in an interview.
They invent capabilities that did not exist before. The first six months are about making the team faster at work they already do. The next six are about asking what was impossible before agents and shipping the first version. Real-time brand sentiment monitoring across thousands of AI conversations. Competitive displacement narratives that deploy within hours of a competitor move. Answer engine optimization systems that win citations across ChatGPT, Perplexity, Claude, and Google AI Overviews. The marketing engineer is the person who sees the new category of work and ships the first version while everyone else is still scoping the RFP.
They make everyone else faster. A content lead ships more, faster, better. A PMM runs a launch analysis in a morning instead of a week. A demand gen manager gets the dashboard on Monday instead of never. The leverage shows up in other people’s output, which is why a good one is worth three hires.
They are measured like marketers, not like engineers. Pipeline. Reach. Conversion. Velocity. What they build has to move those numbers. If it does not, it should not have been built.
What they do not do: own a channel or manage the agency. Pull them into that work and you scoped the role wrong.
The real bar is creative judgment
Here is the part every hiring manager is about to get wrong.
The technology is not the hard part. The tools are getting easier every month. A smart person can learn Cursor, Lovable, n8n, and Gumloop in a weekend. They can ship a working agent by the end of the week. The floor is lower than it has ever been, and it is dropping fast.
The ceiling is not the technology. The ceiling is judgment.
Knowing what agent to build is the job. Anyone can deploy one. Knowing which one will move the number, which one is a distraction, and which one looks useful but will rot in three months without a human watching it. That is the part you cannot learn in a weekend. That is the part that separates the marketing engineer who compounds from the one who ships a hundred tools nobody uses.
Creativity is not a soft skill in this role. It is the skill. Taste tells you what not to build. Imagination tells you what has not been built yet. Narrative instinct tells you how the output will land with a customer who has never heard of your company. Product sense tells you when a workflow is ready and when it is still a demo. None of that comes from the stack. All of it comes from the person.
The tooling is the paintbrush. The marketing engineer is the artist. A beginner with a great brush still paints a beginner’s painting. Give a great painter a cheap brush and you still get a painting worth hanging. The tools compound talent. They do not replace it. The companies that miss this hire for the tools and wonder why the output feels hollow.
This is why the role is rare. Plenty of marketers are creative. Plenty of operators can build. The overlap is small. A marketing engineer who can ship but cannot decide what to ship becomes a very expensive intern. One who can imagine but cannot build ends up filing tickets with engineering like every other PMM. You need both halves in the same seat, and you need the creative half to lead.
Where they sit
Most teams get this wrong on day one, and the role never recovers.
Report them to the CMO/VP of Marketing. Not a chief of staff. Not the head of ops. Not a dotted line into engineering. The person who owns the GTM number. The role only works when the top of marketing is personally invested in what gets shipped and personally embarrassed when nothing does. Three levels down, the first time a campaign goes sideways, this hire gets pulled into the fire drill and never comes back. I have watched this happen twice. Both times the role was declared a failure within a year. Both times the failure was the reporting line, not the hire.
Keep them as a team of one for the first six months. Not inside marketing ops. Not inside analytics. Not matrixed into demand gen. A direct line to the VP and a lateral mandate across the org. Their job is to make everyone else faster, and they cannot do that from inside someone else’s backlog. The moment this role is reporting into a functional leader with a number to hit, the work bends toward that leader’s quarter and the rest of the team stops getting the leverage.
They own what they build. They influence everything else. A marketing engineer who “owns demand gen” becomes a demand gen manager with extra steps. A marketing engineer who builds the tools demand gen uses becomes a force multiplier across every channel at once. The second version compounds. The first one does not. The distinction sounds subtle in the org chart conversation. It is not subtle in year two, when the first version is managing one channel and the second version has made four teams twice as fast.
Close to engineering, not reporting to engineering. A standing line to a friendly engineer for the hard stuff. Infra, data pipelines, anything that touches production systems engineering depends on. Not on the eng team. They move at marketing speed, which is faster than eng speed. That is a feature, not a bug you need to fix by moving them under a staff engineer.
One more thing that matters more than it should. The VP of Marketing/CMO needs to be the one who fights for the comp band, writes the req, runs the interview loop, and defends the role in the first operating review where someone questions it. Delegating any of that to a recruiter or a chief of staff is how the hire arrives under-leveled, under-paid, and reporting to the wrong person on day one. The top of marketing has to own this hire the same way the CTO owns the first staff engineer. Anything less and the role gets treated like a nice-to-have, which is how it ends up acting like one.
What to pay them
Base of $180,000 to $220,000, plus equity on the senior IC band. Higher in San Francisco or New York, where you are competing directly with senior product engineering and senior PM comp.
If finance pushes back on the band, three arguments.
You are not competing with marketers. You are competing with senior software engineers and senior product managers. The strongest candidates have those offers in hand, often multiple. Anchor the role to marketing comp and you lose the top quartile of the funnel before the first interview. Anchor it to senior IC comp in adjacent functions and you get to choose from the real pool. Every company I have seen try to hire this role at marketing band rates has ended up either reposting the req six months later or settling for the candidate who could not get the senior engineering offer. Neither outcome is cheap.
The ROI math is straightforward. A competent marketing engineer retires or renegotiates vendor contracts in the first six months as a byproduct of the work, not as a goal. In most mid-market marketing stacks, that delta alone covers a meaningful share of the fully loaded cost of the role in year one, before counting any output gain across the rest of the team. Ask your finance partner to model the range against your current martech spend. The conversation gets shorter.
This hire compounds. The value in year two is larger than year one because the tools, workflows, and measurement systems they built are still running and still producing leverage. You are not paying this person to do work. You are paying them to build systems that make future work cheaper. That is a structurally different argument than a typical headcount case. Finance partners find it credible when it is made correctly, and they find it unconvincing when it is made by someone who has not done the modeling. Do the modeling before the conversation.
A note on equity. Compensate this role on the senior IC equity band, not the marketing band. The two are often different at companies of the same size and stage, and the gap is wider than most comp teams realize. If your equity framework treats marketing as a lower-equity function by default, the offer will underperform in the market against the engineering and product offers these candidates are comparing it to. Flag this to the comp team before you write the first offer, not after the first candidate declines.
One more thing. Do not cheap out on the first one. The first marketing engineer you hire sets the internal comp anchor for every one that follows. Underpay the first, and the second is harder to hire, because the band on the job ladder is already set too low. The cost of fixing that later is higher than the cost of paying the first one correctly.
The Job Description (steal this)
Drop this into Greenhouse. Edit the company-specific lines. Ship it.
Marketing Engineer
Reports to: VP, Marketing/CMO
Level: Senior IC
Location: Remote, United States
Compensation: $180,000 to $220,000 base, plus equity on the senior IC band
About the role
You will be the first marketing engineer on this team. You will report to the VP of Marketing/CMO and work directly alongside content, demand generation, and product marketing. Your mandate is simple and hard. Make the entire marketing function measurably faster, more accurate, and more inventive by building the agents, tools, and measurement infrastructure the team runs on.
This is a builder role inside a marketing organization. It is also a creative role. You will spend your days writing prompts, designing systems, and shipping production software with modern AI tools. You will also spend them in narrative meetings, on customer calls, and in positioning debates where you argue for what the work should say. Both halves are the job. Candidates who want to do only one will not succeed here.
You will be held to marketing outcomes, not infrastructure metrics. Pipeline, reach, conversion, velocity, share of voice across the answer engines your buyers now default to (tracked in Profound). The systems you build are the means. The numbers are the end. The creative judgment to know which systems are worth building is the reason we are hiring you and not a junior engineer with a prompt library.
What you will do
Deploy agents as production systems. Real systems with branching logic, model calls, data enrichment, human review, and measurable outputs. Research agents that pull competitor signal every morning. Enrichment agents that keep the CRM clean without a human touching it. Outbound agents that write like a human because a human wrote the pattern once. You own reliability, unit cost, and output quality. When something breaks in production at an inconvenient hour, you are the one who fixes it. This is not a prototyping role.
Build internal tools that retire vendor contracts. An enrichment and scoring system that beats the platform we are about to renew. A content operations workflow that turns one customer interview into a week of publishable assets in under a day. A competitive intelligence tool that surfaces signal from public sources every morning. A first-party attribution view that reflects how our business actually works. An answer engine optimization system built on Profound that wins citations across ChatGPT, Perplexity, Claude, and Google AI Overviews. You choose the stack. The team depends on the output.
Own the marketing measurement layer. Integrate data from the warehouse, the CRM, the product, and the model layer into a system the team actually uses. Build the views the VP opens in every operating review. Surface the metrics that predict revenue. Retire the metrics that do not. When the attribution is lying, you are the person who tells us and brings the query that proves it.
Invent capabilities that did not exist before. The first six months are about making the team faster at work they already do. The next six are about asking what was impossible before agents and shipping the first version. Real-time brand visibility monitoring across thousands of AI conversations using Profound. Competitive displacement narratives that deploy within hours of a competitor move. New categories of work nobody on this team has named yet. The range to see the new category is as important as the skill to build it.
Make the rest of marketing faster through direct partnership. Embed with content, demand generation, and product marketing. Find where each function is blocked by manual work or missing data, and build the specific thing that unblocks them. Your performance will be measured in part by the output velocity of the people around you. If the content lead is not shipping more, faster, better because of what you built, the work is not landing.
Make rigorous build-versus-buy decisions. Build when building is the right answer. Buy when buying is the right answer. The skill we are hiring for is the judgment to distinguish between them, and the taste to know that a working tool nobody uses is worse than no tool at all. A great marketing engineer kills more ideas than they ship. We expect the same from you.
Bring creative judgment to every build. The creative half of this role is as load-bearing as the technical half. You decide which agents are worth building and which are distractions. You shape the voice and point of view of the outputs your systems produce. You read a brief, a landing page, a launch plan, and you know when the idea is sharp and when it is slop. You argue for positioning in the room. You kill work that looks useful but will rot without a human watching it. The tools are the paintbrush. The taste is the job.
What success looks like
At 90 days. You have shipped at least one internal tool the team uses weekly. You have mapped our marketing data infrastructure end to end. You have a written point of view on the three highest-leverage things to build next and the two things we should kill.
At six months. You have replaced or retired at least one underperforming vendor. You have put at least two production agents or workflows into daily use. You have built a measurement system the VP of Marketing references in every operating review. One person on another team can name a specific thing you built that made their week faster.
At twelve months. The marketing team ships materially more output per person than it did before you joined. The rest of the company can name specific things you built and specific ways the work got better because of them. You have a defensible point of view on whether and when to hire the second marketing engineer, and the beginnings of a plan for what that team looks like in year two.
What we are looking for
Creative judgment about what to build. This is the first and most important attribute. You consistently identify which problems are worth solving and which are distractions. You build systems people actually use, not systems that demonstrate technical capability. You can look at a workflow and see the shape of the hundred others that run on the same structure. The tools are the paintbrush. You are the artist. We are hiring the artist.
Demonstrated experience shipping production software with AI tools. At least one substantial project that other people depended on to do their work. Built inside a company, inside a founding team, or independently. We will evaluate the work directly. Show us the thing, not the resume.
Fluency with modern AI tools and workflow systems. You build with the current generation of AI-native tools rather than from scratch. Working proficiency in at least two of Lovable, Replit, Cursor, v0, Bolt, or equivalent. Working proficiency in at least one modern workflow or agent platform such as n8n, Gumloop, Relay, or Zapier with AI actions. Direct hands-on experience with at least two frontier LLM APIs, including evaluating model outputs against each other for real tasks. Working knowledge of Clay or a comparable enrichment tool. Hands-on experience with Profound. You have used it to monitor share of voice across answer engines, identify citation gaps, and ship the work that closes them.
Sound judgment about when to apply AI. You have deployed language models in production and have informed opinions about model selection, prompt design, evaluation, cost management, and failure modes. You know when a model is the right tool and when a simpler approach will outperform it.
Narrative and product instinct. You can tell when an output is sharp and when it is slop. You can read a landing page and know what is off. You can hear a customer problem and see the agent that solves it. This is the creative range the role depends on, and it is the reason we are not hiring a junior engineer.
Strong written communication. You will explain technical work to non-technical stakeholders, on an ongoing basis, in writing. Clarity, concision, and narrative structure are required. If you cannot write, you cannot do this job.
Operating speed consistent with a marketing function. Marketing runs in cycles of days, not sprints. You are comfortable shipping working software in short timeframes, iterating against real feedback, and accepting that some work will be revised or discarded. If your instinct is to spec for two weeks before you build, this is the wrong role.
Compensation and structure
Base salary $180,000 to $220,000, calibrated to depth and scope of demonstrated experience. Equity on the senior IC band, meaningful at this stage of the company. Remote within the United States.
The clean JD formatted for Greenhouse is downloadable here. Steal it, edit the company-specific lines, and ship it.
90-day scorecard
If you cannot tell whether the hire is working by day 90, you scoped the role wrong.
Days 1 to 30. Land.
Mapped the marketing stack end to end. Knows every system, where the data lives, and which connections are brittle. Can draw it on a whiteboard from memory.
Shipped one small thing the team uses. A script, a dashboard, an internal tool. The artifact does not matter. The evidence that they can ship inside this company does.
Built a working relationship with one counterpart in every adjacent function. Engineering, data, PMM, content, demand generation. Each of those people can tell you what the new hire is working on this week.
Written point of view on the three biggest time sinks on the team. Documented. Shared with the VP. Prioritized. The two they plan to kill go in the same doc.
Days 31 to 60. Build.
Shipped at least one production workflow that replaces a manual process. Measurable time saved, quantified in writing. “We got four hours a week back on the content QA loop,” not “the team feels faster.”
Built at least one measurement view that did not exist before. The VP of Marketing opens it in the weekly operating review, unprompted.
Killed, renegotiated, or consolidated at least one vendor contract. Savings documented and returned to the budget or redeployed. The finance partner knows the number.
Public backlog in place. Prioritized and versioned. Anyone on the team can see what is coming, what is in flight, and what was explicitly deprioritized and why.
Days 61 to 90. Compound.
Shipped something that made another marketer measurably faster. Named person, named workflow, quantified time saved. “Sarah on content used to spend six hours a week on competitive research. Now she spends forty minutes.”
Two or more production workflows running with real reliability. Logged, monitored, and debuggable by someone other than the marketing engineer in an emergency. The bus factor is not one.
Presented at a marketing all-hands or operating review. The team understands what was built, why it matters, and what is next. If nobody on the team can explain the role in their own words after this presentation, the role is not landing.
Formed a point of view on the second marketing engineer hire. They are thinking about the shape of the function, not just their own backlog. This is the earliest signal of a senior IC who will eventually lead one.
The bar
Two questions at day 90, asked honestly.
Did the marketing team ship materially more than it would have without this hire?
Is the rest of marketing actively asking for more of what this person does?
If both answers are yes, the hire is working. Invest. Expand scope. Start the case for the second one.
If the answer to either is no, the scorecard above will tell you where the breakdown is. Two failure modes to watch for specifically.
They shipped nothing. The role requires someone who ships under ambiguity. If 90 days produced no working artifact, the candidate does not have that disposition, regardless of how they interviewed. This is not coachable on the timeframe the role operates on.
They shipped a lot but nothing the team uses. The role requires taste about what to build. If 90 days produced ten tools and none of them are in daily use by anyone other than the builder, the candidate is building for themselves, not for the team. This is the creative judgment failure mode from earlier in this piece, arriving exactly where we said it would.
Either failure mode resolved at day 90 is less expensive than the same failure mode resolved at day 180. The cost of not having the conversation at 90 is three more months of budget, three more months of the team not getting leverage, and a role that now has a reputation problem inside the company. Have the conversation. On time.
How to know you hired wrong
Four signals. Any one of them is enough to start the conversation at day 60, not day 90. Waiting until 90 is how you end up in month six explaining to your CEO why the role is not working and why you did not see it coming.
They are waiting for specs. A marketing engineer who needs a PRD for every project is a marketing ops manager with different tools. The role requires going from “this is slow” to a shipped solution without a PM in the middle. If at day 45 they are still asking what to build rather than proposing what to build, the disposition is wrong. This is not coachable in the timeframe the role operates on. You can teach a tool. You cannot teach the instinct to ship without permission.
They are building the wrong size of thing. At day 45 they are six weeks into a platform rebuild rather than shipping small things the team uses. The instinct is inverted. The job is a compounding sequence of small wins, not a single big reveal. The correct pattern in the first 90 days is five small tools in daily use, not one large tool still in development. If you see the inverse, the builder is optimizing for their own portfolio, not for the team.
Nobody on the team can tell you what they are working on. Ask the content lead. Ask the demand gen owner. Ask the PMM. If all three shrug, the role is not landing. A marketing engineer whose work is invisible to the people they are supposed to make faster is working in the wrong direction, regardless of what they are building. The correction is fast and specific. Embed them with one function for two weeks and measure whether adoption changes. If it does not, the fit is wrong and two more weeks will not change that.
They are performing AI instead of using it. Every conversation is the newest model, the latest agent framework, the workflow tool someone posted about on LinkedIn that morning. Nothing has shipped. You hired a hobbyist. The best ones are boring about the technology and precise about the output. A useful diagnostic: ask what they shipped in the last two weeks and what it is producing for the team. If the answer is a demo instead of a deployed system, you have your signal.
A note on the conversation itself
Having this conversation at day 60 is the correct move in every case, even when you are not sure. A senior IC who is landing well will not be destabilized by honest feedback at day 60. A senior IC who is not landing well needs the feedback to have any chance of recovering. The only case where day 60 feedback is wrong is if you are avoiding the conversation because you are uncertain, which is the exact case where avoiding it causes the most damage.
If after day 60 feedback and a clear recovery plan the signals persist at day 90, the answer is not more time. It is a clean separation, handled well, and a restart of the search with sharper filters. The cost of a wrong hire in this role is higher than in most, because the whole team is depending on velocity that is not materializing. Every week you spend waiting is a week the rest of marketing is not getting the leverage the role was supposed to produce. That cost is cumulative. Act accordingly.
The line
The first marketing engineer you hire will shape the next five.
They will set the bar for who gets hired after them. They will set the expectation for what this role can produce. They will decide, through what they ship and what they refuse to ship, whether the rest of your org treats this as a real discipline or a vanity title the marketing leader added to look modern.
Get the first one right and the function compounds. Every tool they build makes the next hire more productive. Every workflow they deploy lowers the marginal cost of the next campaign. Every measurement view they ship sharpens the judgment of the people around them. Two years in, you are not running the same marketing function with AI stapled on. You are running a different function. The competitors trying to copy it in 2028 will discover that the lead was structural, not tactical, which is the kind of lead that does not close without spending two to three times the comp to pull senior people out of the companies that moved first.
Get the first one wrong and the opposite happens. The tools do not get built. The vendor contracts do not get killed. The team does not get faster. You spend the credibility of the role itself, which makes the second hire harder than the first one should have been, because the story inside your company is now that this role does not work.
This is why the req matters more than the sourcing pass. Why the scorecard matters more than the interview loop. Why defending the comp band matters more than closing the candidate. The cost of getting this hire wrong compounds in the same direction as the value of getting it right, and both curves are steep.
You now have the playbook. Scope the role. Steal the req. Run the scorecard. Defend the comp.
Then go find the person.
About StackedGTM.AI
StackedGTM.ai shapes how modern GTM categories are understood, evaluated, and bought. Playbooks, frameworks, and operator intelligence for the CMOs, VPs, and founders navigating the AI shift.
Written by someone who has run the motion, for the operators running it now. No hype. No cosplay. No fluff.
Wow - literally learned about the role of marketing engineer from Emily Kramer a couple of days ago and now we have a guide <3 excited to dive in, thanks Josh.