0→1 ProductJob Tech · AI · 2025Designer & Builder

GoWork
Hunt smarter.
Not harder.

Designing and building an AI-powered job hunting platform that brings clarity and control to one of the most stressful experiences for early-career professionals.

Role
Founding Designer & Builder
Timeline
Jan 2025 – Present
Stack
Figma · Next.js · OpenAI
Type
Self-initiated
15
User interviews
40
Survey participants
10
Usability tests
4
Features shipped
Background

I built this because I lived the problem.

Job hunting as a student in a new city meant juggling spreadsheets, browser tabs, tailored resumes, and a constant feeling of sending applications into a void. I wasn't alone.

After talking to 15 job seekers and surveying 40 more, the same frustrations surfaced repeatedly. So I built the tool I wished existed — owning the full product lifecycle as designer, researcher, and developer.

“I have a whole Google Sheet just to track where I applied. It still doesn't help.” — Research participant
01

Lost in the process

Most users maintained 2–3 external tools just to track applications — and still felt out of control

02

Resume paralysis

Tailoring per role felt too slow — most sent the same resume everywhere

03

Rejection void

Applications disappeared with no signal, no feedback — no way to improve

04

Tab overload

A single application averaged 5+ open tabs — LinkedIn, company site, resume, cover letter, spreadsheet

01 — Research

The same four problems.
Every single interview.

30–45 minute semi-structured sessions, focused on current workflow and emotional state during the job hunt.

01
Chaos over clarity
Most users maintained 2–3 external tools just to track applications.
I applied somewhere last week but I can't remember if it was on LinkedIn or their site.
02
Resume anxiety
Users knew they should tailor their resume but found it too time-consuming.
I know I should customize it but I just don't have the time.
03
Invisible rejection
Applications disappeared into a void — the most emotionally draining insight.
I just ghost them back at this point. There's no point waiting.
04
Tab overload
A typical session: LinkedIn, company site, resume doc, cover letter, spreadsheet.
My browser looks like I'm losing my mind by the time I finish one application.
02 — The Goal

One place. Full control.

01
Always know the status of every application
02
Know how your resume fits a role before you apply
03
Apply across sites without the repetition and copy-paste

Feature 01

Your entire job hunt, in one view

A centralized dashboard showing every application — company, role, status, date, and next action. No more spreadsheets.

Usability tests showed users checked application status most frequently right after submitting. The dashboard was designed to make that moment feel reassuring, not anxious.
Your entire job hunt, in one view
Feature 02

Know your fit before you apply

Paste a job description and GoWork scores how well your resume matches it — powered by OpenAI. See exactly where you're strong and where you have gaps.

Early testers used the score as a filter — if under 70%, they'd either improve the resume or skip the role entirely.
Know your fit before you apply
Feature 03

Up and running in under 2 minutes

Onboarding was designed to get users to their first tracked application as fast as possible. Upload your resume and you're set.

The version that asked for a resume first had the highest completion rate because it reduced the number of decisions users had to make.
Up and running in under 2 minutes
Feature 04

Apply anywhere, without the copy-paste

A Chrome extension that auto-fills application forms across job sites — synced to your GoWork profile.

The biggest complaint in testing wasn't the tracking — it was the repetition of filling the same fields on every site.
Apply anywhere, without the copy-paste
Key Design Decisions

The tradeoffs that
shaped the product.

Resume upload vs manual input
Resume first, everything else second

Moving the resume upload to step one cut friction immediately and raised onboarding completion.

Resume upload as step one
Score as number vs score as signal
Show the breakdown, not just the number

An 88% match score alone doesn't tell you what to fix. The breakdown gave users something actionable.

Always show the breakdown
Auto-track vs manual logging
Zero-effort tracking via extension

Manual logging was the most common reason users abandoned tracking tools. Auto-tracking fixed that.

Automatic tracking by default
Reflection

What building 0→1
alone actually teaches you.

01
Constraints sharpen decisions

Wearing both designer and developer hats meant every design decision had to justify itself technically.

02
Research earns trust

Every major feature came directly from interview insights. When you can point to a real user frustration, decisions become easier to make.

03
Shipped beats perfect

Having a live product changed how I think about design tradeoffs. Good enough and shipped beats perfect and stuck in Figma.

Next Case Study
Liveasy
UI Redesign · Logistics TMS · Trust-first design
View Case Study →