How One Job Seeker Turned Claude Into a Career Coach and Landed a DoorDash Role

How Gono Built an AI Job-Search Workflow

Margareth Felicia Gono, 24, finished a master's degree in applied analytics in May and recently started as a senior associate at DoorDash. Between graduation and that offer, she ran her job search less like a scattergun and more like a data project — with Anthropic's Claude as the engine.

As an international student who needed visa sponsorship, Gono initially applied broadly. She then started tracking which roles drew responses and which did not. Using Claude, she extracted key skills, keywords and themes from each job description, then compared them against her résumé to see which terms were missing — a do-it-yourself version of the automated screening many employers use. The goal, she said, was not to have AI invent experience but to surface where her real experience was poorly communicated.

She also used Claude as a career coach to identify what she actually enjoyed — talking to people, design, and pattern-finding — and built three résumés targeting analytics, sales and marketing, and consulting. A dashboard tracked the full funnel: 186 applications, 44 cold outreaches, 19 interviews, six take-home assignments, three final rounds and one offer.

In interviews, she asked permission to record when recruiters said they were using AI note-takers. Most declined; where she got a yes, she ran sentiment analysis on the transcripts to spot hesitation, filler words and moments of engagement. She also used Claude to scan interviewers' LinkedIn profiles for shared interests — one connection over a mutual cheese club proved more useful than any cover letter.

What Her Funnel Numbers Reveal About AI-Era Hiring

What the Funnel Numbers Actually Show

Gono's dashboard — 186 applications yielding 19 interviews and one offer — is a roughly 10% interview rate and a 0.5% offer rate. That is not unusual for high-volume, early-career applications, but it is a useful benchmark for anyone calibrating expectations. The more telling figure is the 44 cold outreaches: a channel that often outperforms blind applications because it reaches a human before the screening software does.

AI as Table Stakes, Not Advantage

Gono's own conclusion is the sharpest insight in the story: "Everyone has access to AI now, so that isn't a differentiator." Résumé keyword-matching and interview sentiment analysis are cheap and widely available. The durable edge she describes comes from judgment — knowing which patterns matter, which roles to skip, and how to hold a real conversation. That mirrors a broader shift in white-collar hiring, where AI tools have flattened the tactical layer of job searching and pushed differentiation back toward human signal: referrals, genuine rapport and clear self-knowledge.

The Recording Question

Her practice of asking to record interviews — and being refused by most employers — highlights an unresolved norm. Companies increasingly deploy AI note-takers on their side while candidates are often denied the same capability, an asymmetry that may draw more scrutiny as AI-mediated hiring spreads. For context, several US states require all-party consent for recording conversations, which likely explains some refusals regardless of company policy.

What This Means for Employers

Candidates like Gono are arriving at interviews having already analyzed the job description, the interviewer's background and their own past performance. Recruiters who rely on generic screening may find that the most prepared candidates are also the ones most likely to detect a poorly defined role — and to walk away from it.

Practical Lessons for Job Seekers Using AI

AI tools can tighten a job search, but they cannot substitute for judgment about which roles are worth pursuing.

• Build a simple funnel tracker (applications, outreaches, interviews, offers) so setbacks read as conversion rates rather than personal failures — Gono's 186-to-1 ratio is a realistic baseline, not an anomaly.
• Use AI to compare your résumé against each job description for missing keywords, but change only bullets that genuinely misrepresent your experience; do not let it invent skills.
• Prioritize cold outreach to specific people over additional blind applications — Gono's 44 outreaches were a distinct channel from her 186 applications.
• If you want to record interviews, ask explicitly and expect refusals; in all-party-consent jurisdictions, recording without permission is legally risky.
• Research interviewers' public profiles for genuine shared interests, but only raise them if the connection is real — the cheese-club moment worked because it was authentic.
• Prepare two or three questions about the interviewer's own career path; Gono credits this with shifting interviews from auditions to conversations.