Skip to content

Portfolio

Personal projectWorking projectAI workflow platform

Orqen

A platform for building AI workflows visually, connecting Google Gemini, external APIs, conditions, database writes and human approval steps.

The challenge

A single AI prompt rarely does a whole job. Real work needs its output checked, routed to other systems and sometimes approved by a person before anything happens.

The approach

Model each workflow as a directed graph, run it on the server rather than in the browser, and enforce each organisation's boundaries and roles in the database as well as in the server functions.

What we did

Abrar built the workflow editor, the execution functions and the access rules, working to a written technical brief. All commits in the public repository are Abrar's.

The solution

Users drag steps onto a canvas and start a run. Runs pause at approval gates until an Editor or Owner approves, and live GraphQL subscriptions show progress without polling.

Key features

  • Drag-and-drop workflow editor built with React Flow
  • LLM, HTTP request, condition, database write and approval steps
  • Approval gates that pause a run until an authorised person responds
  • Live run monitoring through GraphQL subscriptions
  • Separate organisations with Owner, Editor and Viewer roles
  • Row-level security in PostgreSQL through Hasura

Where it stands

The source code is public, and the project runs on a Nhost backend you set up yourself. There is no hosted demo, and no client results are claimed.

Have a project in mind?

Tell us what you need. The first 30-minute call is free, and we'll talk it through with you and give an honest view of whether we are the right fit. We reply within 24 hours, usually sooner.

Not ready to talk yet? Build a project brief first