Apply.ai — Finish-Up-A-Thon Submission

Kaleab Teweldbrhan — Thu, 28 May 2026 07:48:34 +0000

This is a submission for the GitHub Finish-Up-A-Thon Challenge

What I Built

Apply.ai is a job-application workspace platform that helps candidates stop repeating resume work for every single job application.

The idea started from my own frustration while job hunting. I got tired of constantly editing, uploading, and rewriting my CV for every role I wanted to apply for. Every application felt repetitive:

Upload CV
Rewrite achievements
Tailor for the role
Repeat again for the next job

So I built Apply.ai to automate and simplify that entire workflow.

Apply.ai pulls data from GitHub, LinkedIn, resume files, and custom sources once, then turns everything into a reusable portfolio. From there, users can create job-specific workspaces where they tailor applications without rebuilding their CV from scratch every time.

The platform is built around a simple flow:

Sources → Portfolio → Workspace

Key Features

Generate ATS-friendly, job-specific CVs tailored to each role.
Create shareable CV links instead of only sending static PDF resumes, with full customization support.
Generate personalized materials like cover letters, “Tell me about yourself” answers, and interest statements using relevant experience from the user’s portfolio.
Let recruiters chat directly with the candidate’s CV and portfolio context instead of only reading static documents.
Help recruiters quickly evaluate candidate fit against job requirements using a recruiter-style AI agent that compares the role requirements with the CV.
Provide recruiter-style feedback while tailoring a CV, showing weaknesses, gaps, and suggestions for improvement before applying.

The goal is to make job applications faster, cleaner, and less repetitive while helping candidates present themselves better and communicate their real value more effectively.

Demo

Live App: https://apply.kaleab.dev

Product Walkthrough:https://app.supademo.com/demo/cmo8mcata1w4ml2dy8hh353ni

The Comeback Story

This project started when I was actively looking for jobs and realized how much time I was wasting repeating the same resume workflow over and over.

I had an early version of the idea, but it was unfinished and missing structure. During the Finish-Up-A-Thon, I focused on turning it into a real usable product.

I improved:

The overall architecture
Portfolio and workspace flows
Source integrations
Resume tailoring workflows
Recruiter-facing sharing experience
UI consistency and product direction

The biggest change was shifting the product from “AI resume editing” into a full job-application workflow system.

Now instead of treating every application like a completely new task, users can reuse their career data and tailor applications much faster.

My Experience with GitHub Copilot

GitHub Copilot helped me move much faster during development.

I set up the basic architecture and product direction, then let Copilot handle a lot of the heavy lifting while building features. It was especially useful for:

Generating repetitive boilerplate
Refactoring components
Speeding up UI implementation
Helping wire together application logic
Exploring faster implementation ideas

Instead of spending time on repetitive coding tasks, I could focus more on product thinking, workflows, and improving the user experience.

Copilot felt less like autocomplete and more like having an assistant helping me maintain momentum while building.

DEV Community: Kaleab Teweldbrhan

Apply.ai — Finish-Up-A-Thon Submission

What I Built

Demo

The Comeback Story

My Experience with GitHub Copilot