Substack Pays Its Product Managers $1.42 Million. Its Designers $199,000.

Where the Money's Going

Substack has 17 roles open right now, the visible edge of a $100 million Series C closed in July 2025, Substack.com/jobs reported, with a mandate to scale the infrastructure that lets writers own their audience and revenue. The roles posted since the raise reveal where that capital flows. Engineering carries four of the 17 openings. Sponsorships holds three. Operations claims three. Communications has two. Partnerships has two. Marketing, Product, and Standards & Enforcement round out the list with one apiece. The tilt is unmistakable: revenue and trust (sponsorship tooling, enforcement, operational scaffolding), not core publishing features.

Location tells the same story. Eight roles list San Francisco (HQ) as a primary or optional location. Eight list New York. One sits in London (Events Lead, UK & International, remote). One sits in Washington, D.C. (Head of News & Politics Partnerships, remote). Four carry a remote designation alongside hub options. Fifteen of seventeen are hybrid; two are fully remote. The footprint reflects a company still oriented around two physical hubs but building a distributed bench for specialized functions.

Engineering's four openings cover community, growth, and sponsorships product surfaces: two in sponsorships, one in community, one in growth. Operations' three roles are Director of Strategic Business Operations, Head of Strategic FP&A, and IT Systems Administrator. Communications' pair are Head of Events (New York) and Events Lead, UK & International (London, remote). Partnerships' two are Head of Business, Tech & Finance Partnerships (San Francisco) and Head of News & Politics Partnerships (Washington, D.C., remote). The single Product requisition is Product Manager — Growth (San Francisco). The lone Standards & Enforcement seat is Standards & Enforcement Specialist (remote, New York, San Francisco).

Revelio's workforce data puts headcount at roughly 1,634 as of March 2026, more than double the 677 reported in 2023. Finance and Operations now make up 51.5% of the organization. Engineering holds 33.8%. Sales and Marketing sit at 14.7%. The open roles reinforce that composition: every non-engineering department except Product is growing faster than the engineering baseline.

The Ashby board shows 57 active postings across 2026, a 78 percent year-over-year jump. New postings per month climbed from five in 2023 to fourteen this year. Yet WorkAnywhere flags hiring status as "quiet right now": one new role in the last 30 days, the last remote post 29 days ago. The discrepancy suggests a burst model: roles open in cohorts, fill, then pause. Candidates hitting the board today are seeing a cohort's tail end.

Eleven of the seventeen roles lack public title and band data in the table above. The department counts are firm. All 17 specific titles are confirmed on the Ashby board: Head of Creator Marketing (Marketing); Head of Events and Events Lead, UK & International (Communications); the Strategic Business Operations director, Strategic FP&A head, and IT Systems Administrator (Operations); the Business/Tech/Finance and News/Politics Partnerships heads (Partnerships); Product Manager: Growth (Product); Creative Strategist, Implementation Lead, Sales Manager (Sponsorships); Standards & Enforcement Specialist (Standards & Enforcement); plus the four Engineering roles listed in the table. What is confirmed: the company is hiring for revenue infrastructure, trust infrastructure, and the operational layer that connects them.

What the Screen Selects For

Substack's public hiring materials reveal a filter built on three intersecting axes: pragmatic shipping velocity, customer-proximity discipline, and a specific AI fluency that privileges production over theory. The criteria appear consistently across the a16z portfolio jobs page, the a16zbuild announcement, and individual role descriptions.

The a16z portfolio page states the operating philosophy directly: "We aim to take pragmatic approaches to problem solving while shipping high-quality products that allow the writing on Substack to take the spotlight." That sentence functions as a filter. "Pragmatic" and "shipping" appear together; "high-quality" modifies "products," not "processes." The spotlight belongs to writers, not the platform. Candidates who optimize for architectural elegance over writer impact misread the brief.

A second, unusual requirement runs through every role: "Every person at the company participates in customer support. We do this to build empathy with our users and enable us to build better products." This is not a cultural platitude — it is a screening criterion. Support rotation forces engineers, designers, and product managers into direct contact with writer friction. The screen favors candidates who have operated under that constraint, or who treat user feedback as primary data rather than a downstream signal.

The a16zbuild post from May 2026 makes the AI bar explicit for a Product Designer role: "Looking for an engineering-minded designer who has interest & experience in implementing user interfaces in code, and who is deeply capable & familiar with LLMs." Two signals. First, the designer must write production code — not prototypes, not handoff specs. Second, LLM familiarity is table stakes, but "deeply capable & familiar" suggests hands-on integration experience, not prompt-engineering demos. The de-emphasis on language mastery in favor of "software production processes and tools" aligns with the pragmatic shipping mandate: Substack wants people who can move an LLM-backed feature from experiment to shipped product without a dedicated ML platform team holding their hand.

That expectation extends beyond designated AI roles. The job board tags "Artificial Intelligence" on the Standards & Enforcement Specialist and Head of Strategic FP&A roles, and "Machine Learning" on two Full Stack Software Engineer positions (Community and Sponsorships). A trust-and-safety specialist and a finance lead are expected to operate fluently adjacent to ML systems. The screen tests for cross-functional AI literacy: can this candidate evaluate model behavior, understand evaluation pipelines, and collaborate with ML engineers without translation overhead?

A September 2025 YouTube interview with leadership adds a first-principles dimension: "we have to take a sort of a first principles approach and not just you know stuff ads in a thing but ask the question like what would the good version of this be and help build that." The context is differentiating from ad-driven platforms. In hiring terms, this screens for people who reason from user value upward, not from industry defaults downward. A candidate who proposes "industry standard" solutions because they're standard will fail; the screen wants the person who asks what the good version looks like for a writer who owns their audience.

Diversity is framed instrumentally: "We believe that a diverse team will help us build a product that best serves the needs of a wildly diverse ecosystem of writers." This is not a compliance statement — it is a product requirement. Writers on Substack span politics, languages, formats, and audiences. A homogenous team builds blind spots into the product. The screen weights lived experience and perspective diversity as predictive of product coverage.

Finally, the mission language — "the core of Substack is this idea of independence… supported by an audience that's there for them is this like crucial ingredient in a healthy culture in a free society" — functions as an alignment filter. Candidates who cannot articulate why writer independence matters, or who default to platform-centric mental models (maximizing engagement, algorithmic distribution), will not pass the cultural screen. The company makes money only when writers make money; the incentive structure is unusual and the screen tests for genuine comfort with it.

Taken together, the formula reads: ship pragmatically, stay close to writers, write production code around LLMs, reason from first principles, cover the writer ecosystem's diversity, and align with the independence business model. The screen is not looking for the best resume — it is looking for the candidate whose default operating mode matches those six constraints.

Pay, Equity, and the Experience Bar

Substack pays like a late-stage company that still recruits like an early one. Levels.fyi data from September 2026 puts median total compensation across all roles at $335,000, but the spread is violent: the lowest reported package, a Product Designer at $199,000, sits roughly one-seventh of the highest, a Product Manager at $1.42 million. That top figure reflects a PM compensation structure blending high base, heavy equity, and performance multipliers most tech companies reserve for VP-level hires.

The Software Engineer median of $335,000 aligns with the company-wide median, but the same source also lists a $200,000 median for the role (likely reflecting base-only figures or a different percentile cut). Candidates should treat $335K as the realistic total-comp anchor for mid-to-senior ICs. Equity vests on a standard four-year schedule: 25 percent each year, no acceleration clauses publicly documented.

Glassdoor's 15 reviews (4.6/5 stars, 82 percent recommend rate) tell a different story about the gate. Interview difficulty rates 2.9 out of 5 (moderate), but only 15 percent of candidates describe the experience as positive. One reviewer said: "Substack's engineering team is one of the best I've worked with in my career. Talented and humble people who genuinely care about the work they do." The low positivity score suggests the bar is high and the process opaque, not that the team is unpleasant.

Zero G Talent's live board data, updated within the past week, shows 13 salaried roles with a typical band of $122,000–$292,000 (median $260,000).

Current openings cluster in two tiers: senior leadership/strategy roles at $200K (Zero G Talent reported)–$300K (Zero G Talent's figures put the ceiling at $300K) in San Francisco, and IC engineering roles at $140K (according to Zero G Talent)–$260K (Zero G Talent's data shows) across both hubs.

The board band is narrower than Levels.fyi's extremes because it reflects posted ranges for open requisitions, not historical outliers. A candidate negotiating from the $140K floor on an IC role has room; the $260K ceiling matches the board median and sits below the Levels.fyi Software Engineer median, suggesting Substack still has headroom to compete for senior talent without breaking its own structure.

Experience expectations are implicit in the bands. The $140K–$260K engineering roles map to three to seven years of production shipping. The $200K–$300K strategy roles require prior FP&A or BizOps ownership at scale. No entry-level roles appear in either dataset.

For applicants, the takeaway is simple: anchor to the median for your track ($335K engineering, $1.18M product, $162K design), know the vesting clock is straight-line four years, and expect an interview loop that filters for depth over speed. The 15 percent positivity rate on Glassdoor isn't a culture flag — it's a signal that the screen works.

How to Clear the Bar

Substack's interview process averages 29 days from first contact to decision, per 20 user-submitted interviews on Glassdoor. It typically opens with a recruiter screen to align background and interest, followed by a hiring-manager conversation that dataford.io describes as "practical and direct." Candidates who treat this like a standard FAANG loop (LeetCode drills and STAR stories) miss what the company actually screens for: operators who can ship.

The clearest playbook comes from the AI Governance newsletter, which tracks a similarly selective segment ($150K–$400K+ roles, drowning in applications but starving for evidence). Its editor puts it bluntly: "Résumés get you a callback. Portfolios get you an offer." Hiring managers and chief AI officers Google candidates before the interview. "If they find a Substack post and a GitHub repo full of impact assessments, you jump to the shortlist. If they find a certification badge and a job title, you're just another candidate."

That translates to a concrete artifact list. The newsletter recommends publishing ten pieces before you apply: an AI system inventory, a completed impact assessment, a policy-to-controls translation matrix, a hands-on bias audit, an agent policy and charter pair, a vendor due-diligence questionnaire, an incident response playbook, a cross-framework mapping, public thought leadership, and evidence you've facilitated real work. Pair every credential with one of these. "I would run demographic parity and equalized odds tests using Fairlearn on the deployer's population, document findings in the impact assessment, and set a monitoring threshold of 4/5ths approval ratio" is the cited hire-me answer: specific, tool-aware, and measurable.

Technical fluency matters more than credentials. Practice explaining, in under 90 seconds each: the difference between supervised, unsupervised, and reinforcement learning; what a foundation model is and why it matters for governance; fine-tuning versus RAG; what "agentic AI" is and why it creates new governance problems; why AI is probabilistic and what that means for testing and validation. Rehearse three case studies out loud (loan approval, hiring AI, healthcare AI) timed to 12–15 minutes each. Record yourself. Refine.

The interview itself rewards operators, not observers. "The winning candidate here isn't the smartest person in the room," the newsletter editor said. "It's the one who sounds like an operator — someone who has facilitated hard conversations before and has scars to prove it." Three mistakes sink candidates: overweighting credentials (AIGP, CIPP, ISO 42001 lead auditor, useful but none sufficient), speaking in principles instead of mechanisms, and not understanding the business. "AI governance exists to enable safe deployment, not to block it." Candidates who position themselves as "no" people lose to those who say "here's how we get to yes safely."

Ask questions that reveal whether the program is real: What's the current maturity (greenfield, formalizing, or scaling)? Who owns AI risk today: CIO, CISO, GC, CRO, or a dedicated CAIO? Where does this role sit and who does it report to? What's the biggest governance gap you're trying to close in six months? Do you have an AI inventory and how is it maintained? How are you preparing for EU AI Act enforcement? What's your incident response process when an AI system fails?

On the application side, one job seeker documented testing AI agents at scale: 819 applications in a month, roughly $50 and two days of setup, yielding five interviews, a 0.6 percent interview rate, or one interview per 200 applications. The agent tailored CVs and cover letters to each job description, filled demographic and motivation fields, and ran in a "review-then-auto-apply" hybrid mode. "I chose the middle ground," the author wrote. Reviewing early applications trained the agent; periodic spot-checks kept hallucinations in check. The side benefit: surfacing companies the candidate actually liked, then pursuing them through traditional channels, which produced the interviews that mattered.

Substack's own AI philosophy signals what it values in builders. The company's 2023 audio-tool launch framed AI as "super-powers" for creators, not replacement. "We've done this all using cutting-edge AI tools, and it reflects our philosophy of not thinking that this AI stuff is ever going to take the place of work done by writers and creators — but instead we think it can give writers and creators super-powers." That same stance appears in its hiring: show you can wield the tools to amplify judgment, not outsource it.

Reddit users have flagged Substack's new AI-detection system for false positives on decade-old human writing. "AI detection systems are notoriously trash. They falsely flag people's work all the time." The takeaway for applicants: don't sanitize your voice into something a detector expects. Write like a person who thinks. The screen is built for that, and the writers who own this platform will spot the difference.

Working in AI? Zero G Talent tracks the openings: see every open Substack role, browse AI jobs, the companies hiring, and the people building the field.