All projects

Full Stack

UK Clearing Helper

A university search tool that solved a problem I experienced myself.

Role
Product, scraper and full-stack development.
Status
Seasonal. Clearing has closed for the current cycle.
Year
2026
UK Clearing Helper landing page with the headline 'Clearing, without the panic' over a photograph of Durham Castle.
Landing page. Hero photo: Durham Castle by Christophe Meneboeuf (CC BY-SA 4.0).

01

The problem

Clearing is stressful and the information is scattered across many university sites, with no single live view of what is open. I lived through it myself, so I built the tool I wanted: one place to see what is open, matched against my grades, with a call script ready before I dialled.

02

My contribution

I built both halves: the Playwright scraper that collects vacancy data, and the Next.js and Supabase app that presents it, stores your grades and target courses, and generates a call script.

It helped me during Clearing and my move to the University of Birmingham.

  • Next.js
  • TypeScript
  • Playwright
  • Supabase

03

How it works

  1. Collect

    A Playwright scraper reads vacancy pages and writes them to Supabase.

  2. Browse

    Every tracked university is browsable with vacancy status and subject filters, signed in or not.

  3. Match

    Save your grades, top three target universities and backup courses, and match them against what is actually open.

  4. Prepare

    Generate a short call script covering who you are, your grades and a fallback ask if the course is full. Edit it and save it per university.

04

Engineering details

  • No UCAS login, ever

    Access is never gated on UCAS status. The UCAS status on a profile is a manual toggle, with no UCAS API integration.

  • Row-level security

    Universities and courses are public-read, and each profile is only readable and writable by its owner. Writes to vacancy data come from the scraper using the service role key, never from the browser's anon key.

  • Scraping that admits its limits

    University sites differ and change, so a scraper is only as reliable as the pages it reads. I measured it rather than assuming: roughly 75% reliable in testing.

05

Results & limitations

It did the job it was built for: I used it myself during UCAS Clearing, and it helped me secure my place at the University of Birmingham.

  • ~70%

    of Russell Group universities covered

    In my own testing during the Clearing period. Not a claim about current coverage.

  • ~75%

    scraper success rate in testing

    From the same testing period.

Limitations and honest caveats

  • Clearing has closed for the current cycle, so out of season the live site shows no vacancies. Hotline numbers and links remain.
  • Vacancy data came from a scraper with partial, varying coverage, so listings could be incomplete or out of date.
  • There is no UCAS integration.