This RevOps Hire Built His Company's Entire GTM Stack Before Hiring a Single Rep

Mark Hardy is the entire Revenue Operations function at depthfirst, an AI security startup — and he built the company’s go-to-market infrastructure before it hired a single sales rep. A September 9 Clay customer profile details how Hardy identifies buyers in two segments — companies replacing legacy tools like Snyk and Veracode, and AI-native dev shops overwhelmed by the code volume tools like Cursor now generate — by layering technographic data, breach and incident signals, 10-K cyber-risk disclosures, and Gong call-transcript mining into Clay Audiences segmented by role, industry, and funnel stage. Rather than fear-based outreach right after a breach hits the news, depthfirst warms prospects first with LinkedIn ads built on its own security research, only initiating direct sales outreach once engagement signals appear.

The tooling story extends past prospecting. Hardy pairs Clay with Claude Code for internal build work — pricing calculators, account-specific landing pages, onboarding materials — and used it to map his entire Clay workspace and audit Salesforce sync, discovering that field population had quietly dropped to 10% before restoring it to 89%. Claude Code also mined Gong transcripts for recurring themes Hardy could feed back into segmentation. His framing of the skill floor for this kind of work is blunt: “you’ve got to download something or get a license… you’ve got to try it” — positioning GTM engineering as accessible through natural-language prompting rather than requiring a formal technical background.

Hardy’s build mirrors what Clay’s own Head of GTM Engineering, Osman Sheikhnureldin, described in a separate profile: his team prototyped an “account agents” tool over a Christmas break, ahead of Clay’s own product organization shipping the equivalent as an official feature — internal experimentation surfacing requirements, like source citations on agent outputs, that later became shipped product. Both stories point at the same shift: a single motivated operator with Clay and an AI coding agent can now stand up GTM infrastructure that used to require a multi-person RevOps build-out, and internal prototypes built to solve one company’s problem are increasingly previewing what vendors ship next.