Maggy

Job seekers applying to multiple roles typically keep one generic resume and manually rewrite it for each posting, or skip tailoring entirely. This means copy-pasting job descriptions between apps, guessing which keywords an ATS is looking for, and manually tracking which roles were applied to at what stage. Without tailoring, resumes may miss role-specific keywords the job description calls for, and without a tracking system, saved or applied roles can quietly stall with no clear next action.

Maggy is an iPhone app where users keep one master resume and share any job posting link into the app from the share sheet, whether from LinkedIn, Indeed, Glassdoor, Seek, or company career pages on Greenhouse, Lever, or Workday. Maggy reads the job description from the link and rewrites the resume against it, showing every edit for the user to review before anything is exported as a PDF. The app compares the tailored resume against the job description to show ATS keyword coverage, gives each saved job a resume score and action plan, and can generate a cover letter from the resume, a base cover letter, and the job description. A feed lets users swipe through matched roles to save or pass, and a tracking view shows each saved job's status (saved, ready, applied, interview) with its next action. The app states it does not fabricate experience or numbers not already in the user's master resume, and resume data is not used to train Maggy's models.

Hiring has turned into an AI arms race, and job seekers are losing it. Employers now screen applicants with applicant tracking systems (ATS) and AI filters that rank people by how well their resume matches each job ad. At the same time, one-click apply has flooded every listing, so a generic resume gets filtered out before anyone reads it. Tailoring your resume to each job is now the only thing that works, but doing it by hand takes 30-60 minutes per application, so most people stop. Until recently, doing this well with software was either too expensive or too unreliable. Today's AI models can finally read a job ad, understand someone's real experience and rewrite it accurately in seconds, at a cost low enough for a consumer app. ChatGPT has also taught job seekers to expect AI help, but they're tired of writing prompts, copying and pasting, and fixing the formatting themselves. That gap between "AI can do this" and "a tool that just does it, on your phone" is open right now.

1. Tailoring without making things up. It's easy to get an AI to write a better-sounding resume. The hard part is matching every keyword to the job while only reframing what the person has actually done. One invented skill destroys trust and can cost someone the job. We show exactly which bullets changed so users can check every edit. 2. Messy inputs. Resumes arrive as PDFs and Word files in endless layouts, and job ads come from many boards in different formats. Reliably pulling out structured experience and real requirements from both is a big engineering problem. 3. Output that passes the filter and the human. The finished resume has to be readable by ATS software and also look good to a recruiter, as a clean, formatted file people can export straight from their phone. 4. Job data at scale. A useful match feed means continuously collecting and cleaning listings from boards like SEEK and Jora, plus individual employer career sites, and keeping that pipeline fast and affordable. 5. Unit economics. Every tailored resume costs us AI compute. Pricing has to cover that cost while staying affordable for someone who is between jobs. That's why we use credits and a Job Hunt Pass instead of an open-ended subscription. 6. Standing out from "just use ChatGPT." The product has to be clearly faster and more trustworthy than a free chatbot, through the saved master resume, visible changes and ready-to-send output.