Resup.dev
a résumé builder aimed at the software that reads it first
- Client
- Own product
- Role
- Personal project
- Duration
- ongoing since 2025
- Delivered
- 31 Mar 2026
- Status
- live
Outcomes
2
input formats accepted
Free-form entry, or an existing résumé parsed and restructured.
parse → structure → optimise → export
pipeline stages
unknown
measured effect on interview rate
Deliberately stated. I have no data, so I claim none.
The problem
Most résumés are rejected by software before a person sees them. Applicant tracking systems parse a document, and a layout that looks considered to a human — columns, tables, icons, a header graphic — is frequently the reason the parse fails and the application disappears. Candidates are optimising for the wrong reader and getting no feedback about it.
Constraints
- The output has to survive parsing, which rules out most of what makes a résumé look designed
- It has to accept an existing résumé as input, not just a form, because that is what people actually have
- Parsing arbitrary user-uploaded résumés means handling documents built every possible wrong way
- Built solo, alongside a full-time job
What I did
- Built the platform end to end: it takes free-form input or an existing résumé and produces a document a parser can read, with templates and export options
- Engineered the résumé parsing and optimisation pipelines in Python using NLP and ML techniques, covering both structural formatting and keyword suitability for applicant tracking systems
- Designed the templates against what parsers can read rather than what looks impressive, which is a genuinely constraining brief
The hard part
Parsing résumés people already have. There is no schema — a résumé is a visual convention, not a format — so a two-column layout, a table used for alignment, and dates written six different ways all have to resolve into the same structure. And the tool has to fail visibly rather than silently: a parser that quietly drops an entire employment section produces a confident, polished, incomplete document, which is worse than refusing the file.
Outcome
Resup.dev is live at resup.dev. It parses an uploaded résumé or takes free-form input, and what comes out is a document I have watched parse cleanly — which is the only claim I can actually make for it.
What this did not do
- I have no data on whether it improves interview rates. I can show that output parses cleanly; whether that converts is unmeasured, and I am not going to imply otherwise.
- Applicant tracking systems are opaque and numerous. Output is built against how parsers generally behave, not certified against any specific vendor.
- Parsing accuracy on unusual layouts is still the weakest part, and heavily designed résumés are where it struggles most.
- It is a side project, maintained in evenings. Uptime is best-effort.
Stack
Screens
A résumé is a visual convention, not a format. Parsing it is the hard part.
Further reading
Related results
Results
- viditparashar.meviditparashar.me › projects › clinical-agents
Clinical Agent Suite — four AI agents for the parts of care that are…
Case studyShipped2025
31 Aug 2025 — … pilots, not asserted. Designed and built four separate agents — medical scribe, patient intake, clinical decision support, and medical coding — on Next.js, LangChain…
- 40%physician documentation time saved
- 4clinical agents designed and shipped
- 2academic health systems integrated
- Next.js
- TypeScript
- Python
- LangChain
- LLMs
- Node.js
- +2 more
- viditparashar.meviditparashar.me › projects › medscribe
MedScribe — an ambient scribe that turns a consultation into a note an…
Case studyLive2025
31 Dec 2025 — … hospitals. Seed-stage: every architectural decision made once and lived with. Built the core pipeline end to end — real-time speech-to-text through medical NLP into a…
- 2hospital chain clients deployed to
- 3clinical document types generated
- 2 monthsfrom empty repo to hospital deployment
- Next.js
- TypeScript
- Node.js
- Python
- LangChain
- LLMs
- +2 more
- viditparashar.meviditparashar.me › experience › hsi-labs
Software Engineer at HSI Labs (Aug 2024 – Aug 2025)
Full-timeAug 2024 – Aug 2025
31 Aug 2025 — Designed and built a suite of AI healthcare agents — medical scribe, patient intake, clinical decision support and medical coding — and the NLP pipelines behind them.…