The Profile That Clears the Bar
Artisan sells AI agents that replace sales development representatives. It employs 35 humans to build, sell, and support them. The contradiction is the product.
Artisan's small size and concentrated hiring profile, with every open role carrying a staff-level title and median pay of $213,000, create a process where a narrow set of traits and preparation steps determine whether a candidate advances. This guide maps exactly who passes, what the work pays, and how each stage works.
The company sits in the AI SDR category, a crowded space where first-generation products earned a reputation for hallucinated emails and low response rates. Artisan's flagship agent, Ava, now hallucinates roughly one in 10,000 emails after a year of work with Anthropic on tighter prompting. Two more agents, Aaron for inbound messages and Aria for meeting management, are slated for late 2025. The roadmap demands engineers who can ship reliable agentic systems, not researchers chasing benchmarks.
Zero G Talent reported six active role families, every one carrying a staff-level title: Staff Software Engineer (Backend), Staff DevOps Engineer, Staff Frontend Engineer, Staff Data Engineer, Forward Deployed Engineer, and a generalist "Builder" track. Salary bands run $130,000–$300,000 with a $213,000 median. Equity is included but not priced against a public valuation. Locations cluster in San Francisco and New York with U.S. remote eligibility for most roles.
The hiring bar reflects a team that has already cycled through painful product lessons. Carmichael-Jack describes the Y Combinator-era version of Ava as producing emails that made him "cringe in pain" — extreme hallucinations, unqualified customers, churn. The fix required rigid prompt architectures, a proprietary brick-and-mortar business database, and the discipline to turn away entire verticals such as offshore development agencies that don't work with agentic outbound. About 250 customers and $5 million in annual recurring revenue came from qualifying heavily, not selling broadly.
That history filters for a specific profile. The CTO hire, Ming Li, previously at Deel, Rippling, TikTok, and Google, signals an emphasis on scaling infrastructure and reliability over pure research. Forward Deployed Engineers sit at the customer interface, translating messy real-world sales workflows into agent configurations. Builders operate across the stack with high autonomy. Staff engineers own entire subsystems: backend reliability, data pipelines, frontend performance, DevOps maturity.
Customer qualification discipline is the other filter Carmichael-Jack has named explicitly. "We've historically sold to a lot of the wrong customers, and learned the hard way that it's not just like a typical SaaS product where you can sell to everyone — you have to actually qualify pretty heavily." Artisan now turns away entire categories, such as offshore development agencies, and pilots success-based pricing where they only charge if the customer gets value. That posture requires sales and product people who can say no to revenue, and engineers who build instrumentation that makes value visible.
That database Carmichael-Jack cites as Artisan's edge also shapes the profile. It's not a generic web index; it's a curated, structured asset that the AI agents query directly. Maintaining and expanding that dataset takes people who understand data lineage, entity resolution, and the difference between coverage and accuracy — skills that don't appear on a standard backend resume.
Remote eligibility on most roles means the team operates asynchronously by default. The forward-deployed engineer role, which sits between product and customer, signals that Artisan expects engineers to work directly with users, not through a product-manager translation layer. That only works if the engineer can communicate, prioritize, and ship without daily standups.
The "Stop Hiring Humans" campaign generated death threats and an April Fool's "resignation" stunt from the CEO. Surviving that noise and continuing to hire 20 people into a 35-person team after a $25 million Series A requires low ego, high noise tolerance, and the ability to separate signal from theater. Carmichael-Jack's operating principle: "Human labor becomes more valuable when you have the AI content." People who thrive here treat AI as a tool, not replacement, and build systems where the human-in-the-loop is the competitive advantage.
Self-assessment: if you want to publish papers, this isn't the place. If you want to own a piece of a production agent platform that 250 companies pay for, and you can demonstrate shipped systems in Python or TypeScript, distributed infrastructure, or sales-tool integrations — the conversation starts. The bar is staff-level contribution from day one.
What the Roles Pay
Artisan's compensation data tells a clear story: a senior-heavy engineering team paying at the top of the market for specialized talent. Board data shows 14 salaried roles with a typical band of $123,000–$279,000 and a median of $213,000. Every role currently listed carries a "Staff" or equivalent title, with no junior or mid-level slots. That concentration shapes the numbers.
The data shows 14 salaried roles with a typical band of $123,000–$279,000 and a median of $213,000. According to Zero G Talent, that concentration shapes the numbers.
| Role | Salary Range (USD/year) | Locations |
|---|---|---|
| Staff Software Engineer (Backend) | $200,000 – $300,000 | San Francisco / New York / Remote (US) |
| Staff Data Engineer | $200,000 – $270,000 | New York / San Francisco / Remote (US) |
| Staff Frontend Engineer | $175,000 – $275,000 | San Francisco / New York |
| Staff DevOps Engineer | $170,000 – $280,000 | San Francisco / New York |
| Forward Deployed Engineer | $175,000 – $250,000 | New York / San Francisco / Remote (US) |
| Builder | $130,000 – $230,000 | San Francisco / New York / Remote (US) |
Four roles explicitly list "Remote (US)" — Backend, Data, Forward Deployed, and Builder. Staff Frontend Engineer and Staff DevOps Engineer show only the two physical hubs. That pattern suggests remote eligibility is role-dependent, not a blanket policy. Candidates should read each posting's location line literally.
The two hubs function as coordination anchors. San Francisco serves as the product and engineering center of gravity: Staff Backend, Staff Frontend, and Builder roles all list it first, and the Forward Deployed role anchors there. New York serves as the commercial and data-engineering counterweight. The Staff Data Engineer role lists New York first, then San Francisco, then Remote (US), a sequencing that mirrors where the go-to-market motion and the data pipelines that feed it are managed. Forward Deployed Engineers also sit in New York. The board data shows a $175,000–$250,000 band for that role, while the San Francisco backend band is $200,000–$300,000, indicating the company prices senior IC work by scope.
Salary bands, ranging from $130,000 for a Builder to $300,000 for a Staff Backend Engineer, are narrow enough that internal transfers between hubs don't create compensation friction. At roughly 35 people currently, with plans to add about 20 more, the entire company can still fit in one large conference room for an all-hands.
Equity appears in Artisan Partners' public benefits summary as "equity or equity-linked incentives" alongside competitive salaries and variable incentives. That language comes from the investment-management firm Artisan Partners, not the AI company whose roles populate the board above. The two share a name but are distinct entities. Zero G Talent's board postings for the AI Artisan do not disclose equity grants, strike prices, or vesting schedules. Until the company publishes an offer-level breakdown or a candidate shares a signed offer, treat equity as undisclosed for these roles. The same applies to variable incentive targets; the board shows base bands only.
Benefits context from Artisan Partners' careers page describes a 401(k) with a 100% dollar-for-dollar match up to the IRS limit, a healthcare plan the firm calls "nationally leading" covering dependents, spouses, domestic and same-sex partners, unlimited annual mental-health visits, a benefit concierge, generous time-away policies, and family-care benefits including fertility treatment, paid parental leave for all parent types, and gender-transition support. Again, this is the investment firm's published program. The AI Artisan has not published a comparable benefits summary on its careers site. Candidates should ask for the current benefits guide during the offer stage rather than assume parity.
The upshot: base compensation for posted roles clusters in the $170,000–$300,000 band for staff-level engineers, with remote eligibility on roughly two-thirds of the slots. Equity, bonus targets, and the full benefits package remain undocumented in public sources. Bring those questions to the recruiter screen, as they are the only forum where the numbers become concrete.
Inside the Interview Loop
Artisan runs its hiring on Ashby. Applications go through the platform, and recruiter follow-ups flow through it too, so the first touchpoint is a structured form, not an email inbox. Scoutify, which tracks the company's public interview loop, notes that auto-apply tools currently cover Greenhouse, Workday, and SmartRecruiters; for Ashby boards like Artisan's, it alerts candidates in real time so they can be among the first applications in. That timing matters: the company sits at roughly 35 people and has that hiring target, per TechCrunch's April 2025 reporting, so each requisition draws a concentrated pool.
The loop Scoutify documents for Artisan is a five-stage sequence built for machine-learning and engineering roles. Stage one: a 30- to 45-minute recruiter or hiring-manager screen covering background, research interests, and the role; ML candidates often get a quick technical discussion about past work. Stage two: a 60-minute technical screen covering coding plus ML concepts, sometimes implementing an algorithm from scratch or debating model-architecture choices. Stage three, the ML Deep Dive, runs another 60 minutes on model evaluation, feature engineering, and production ML challenges. Stage four: a 60-minute system-design or ML-architecture session where candidates design an end-to-end ML system including data pipeline, feature store, training, serving, and monitoring. Stage five, where applicable, is a 45- to 60-minute research discussion or paper review testing depth of understanding and communication of complex ideas.
Not every role hits all five stages. The first-party board data from Zero G Talent shows current openings for the roles outlined above, roles that lean toward production engineering and deployment rather than pure research. For those, the loop typically compresses: recruiter screen, technical screen, system design, and a hiring-manager conversation. The research-discussion stage appears reserved for Applied AI or research-track candidates.
The Ones Who Stay
Artisan's hiring record points to a specific profile: senior builders who have shipped product in messy, high-velocity environments and can operate without guardrails. The company's board listings tell the story directly — the pattern persists. Median board salary: $213,000. This is not a junior training ground.
The CEO's own recounting of the company's first year reinforces what the org chart implies. Carmichael-Jack described the YC-era product as something that had that effect: "extremely bad hallucinations" shipped to real customers. The fix took months of work with Anthropic to build rigid prompt architectures that cut hallucination rates to that level. That trajectory of public failure, technical depth, and measurable recovery selects for engineers who treat reliability as a craft problem, not a compliance checkbox.
If you need structure, mentorship programs, or a well-defined career ladder, this is the wrong place. If you've led a migration that broke production at 2 a.m., fixed it before breakfast, and documented the postmortem so the next person doesn't repeat it, Artisan's door is open.
Working in frontier tech? Zero G Talent tracks the openings: see every open Artisan role, browse frontier tech jobs, the companies hiring, and the people building the field.