Backend · APIAug–Oct 2026 · Internship · Team of 2

Usage Data Pipeline

Usage data that loads itself.

A Node.js and Express service that pulls usage reports from vendor APIs, matches every row to the right record with a multi-tier engine and feeds a dashboard, replacing a spreadsheet rebuilt by hand with one button.

Skills it shows

usage data pipeline · dashboardIllustrative mock

Usage dashboard

Synced 12 s ago · sample data

Load data

214

matched

to review

data gaps

Title ATier 1High

Title BTier 2High

Title CTier 3Medium

Title DReviewFlagged

An illustrative mock with sample data. The real dashboard runs on private records.

RoleFront end and the Node.js service

TeamTwo developers, working Agile

WhenAug–Oct 2026, internship at The Walbrook Institute London

Built withNode.js, Express, Chart.js

Skills it shows

API integration

Pulls COUNTER R5 usage reports from vendor REST APIs on demand, with every endpoint tested in Postman first.

Data matching

Matches on ISBN, title, year and publisher across four tiers, strongest evidence first. Anything ambiguous is flagged for a person, never guessed.

Reliable backend

API keys stay off the client, a second sync is refused while one is running, and vendor errors become plain-English messages.

Professional delivery

Feature branches and reviewed pull requests. It passed acceptance testing and went live.

Problem, build, result

1Problem

The dashboard worked, but feeding it meant rebuilding a spreadsheet from several exports by hand every time.

2What I built

A service between the browser and the vendor APIs that collects, matches, checks and shows: charts, a filterable table and a PDF report from the same data.

3Result

Deployed after passing acceptance testing. Fresh data is now one button.

1 buttonreplaces a manual rebuild

4 tiersof record matching

0duplicate runs: a second sync is refused

Liveafter acceptance testing