Hirable.fyi
Digital career advancing tool for job seekers
1 month into our LinkedIn Launch, we've just hit our first 100 users! Many traffic came from LinkedIn, but also from us submitting our AI tool to AI directories, such as https://www.futurepedia.io/tool/hirable, chatgptdemo, aitree.io etc.
Most of these users aren't our target audience, so we are still working on finding the right audience.
… on the marketing side, I’ve launched on LinkedIn, added Analytics tracking (GA and Vercel), promoted the site on many AI product directories, gotten feedback from potential users, and last but not least, planned a vision forward.
https://medium.com/@oat-on-a-boat/page-11-launch-marketing-feedback-and-next-steps-ae7566372c6e
I deployed hirable on Vercel. It is a great and go-to deployment platform for Next.js, built by the founders of Next.js themselves.
I used NextAuth.js for authentication, with support for Google provider for now. The auth framework was built for Next.js as well which made it very easy to install.
I chose a non-relational MongoDB database to store user sessions and their resume/cover letters, and moved away from localstorage . It works very well with Next.Js auth for authentication. Simply create a mongo DB account, create a new cluster, pick a cluster type and go.
https://medium.com/@oat-on-a-boat/page-10-technical-wrap-up-a-ee0d4a5372a5
Sharing my 3 days of work to build the UI for Hirable.
Stack
Next.js for full-stack React framework. Using the beta version 13 with React Server Components
React
Typescript
OpenAI API for generative AI assistant
TailwindCSS for styling
Vercel for deployment
https://oat-on-a-boat.medium.com/page-7-building-a-web-application-aa5ee7a15448
Our natural next step is to design. We will be using Figma, leveraging some AI-powered plugins to help with the color palette, image generation, and general best practice analysis.
Let’s sketch out a lo-fi draft of what some pages look like, starting with the onboarding. Users can start their journey in 3 different ways:
- From scratch — manual input
- Import from LinkedIn — Important to get this right. It is well structured so we can easily parse the data
- Import from resume PDF — Could be challenging to parse, maybe post MVP
https://oat-on-a-boat.medium.com/page-6-sketching-application-layout-32926d620ed9
Thought deeply about it overnight (til 3 AM) and realized AI is an integral part of the app and should be MVP. Before moving further, we need to validate that we have available resources for AI capabilities that can do what we expect it to do.
What do we want it to do?
- Generate a high-performing resume draft based on a role and general information about the user’s experience/skills
- Rewrite a resume based on a job description
- Rewrite a resume based on other resumes
- Highlight strengths and weaknesses of a resume against that job description, with prompts to improve
- Translate resumes in different languages
- Share insight around critical keywords/outlines that high-performing resumes use for a certain role
https://oat-on-a-boat.medium.com/page-3-validating-ai-feasibility-b81235d94085
https://oat-on-a-boat.medium.com/page-4-validating-ai-feasibility-in-openai-node-js-api-b88c7423f92b
https://oat-on-a-boat.medium.com/page-5-validating-ai-feasibility-in-openai-api-e948931d0db2
Came up with a name for the platform!
The first and most step of a project is, of course, the project name. Who has time to brainstorm different titles and typography in 2023? Let’s use AI.
https://medium.com/@oat-on-a-boat/page-2-name-and-branding-b73cbdd24a12
https://oat-on-a-boat.medium.com/what-i-am-building-2fbfd08ccbb2
I want to build digital resume-building software. Think Facebook/LinkedIn profile but focused on accomplishments/portfolio/references in multimedia format. Choices of different templates and layouts with drag-and-drop UX, and AI-assisted copywriting tools. The AI will also help generate different versions of the resume based on job applications, and help draft cover letters. Applicants can share this link as their portfolio website, or export it as PDF. They can create from scratch or upload their LinkedIn to import from. Resumes can also be in different languages. Added analytics for page views.
Reason being:
- People laid off, want to help
- A lot of smart, talented people have terrible resume-writing skills. They get ignored by companies because of this. This levels the playing field for everyone
- Help people transition from one role to another without drafting another resume
- Easily shared, more analytics on who is viewing. Bring the power back to candidates.
- Resumes should look pretty and presentable. Some folks have no skill, and this hurts their chances
- Pushing the needle to make recruitment more human. You are not hiring a worker to do X for your company. You are hiring a human being that contributes to your culture, introduces game nights, recommends books to colleagues, and has cool music tastes. This data point needs to be evaluated, not just that they can memorize conflict-resolution answers in the STAR method.
- I’ve always loved building beautiful resumes. Started when I was entering tech as self-taught and needed attention. Continued and felt it made me stand out and make a difference to people. Want to do the same for others in a scalable way. Scalable → platform + AI
The next steps are writing and designing. I will use AI for a lot of this. Will continue to write here every time there is an update.
Comment