Roadmap for Resume Matcher

[srbhr]

Planning a roadmap for this project. Let me know your thoughts on the tech-stack.

[by14]

Next JS Frontend [t3 stack]
Fast API python backend
Clerk for auth
OpenAI integration for embedding so we don't need to use GPU [$5 free credit is 15,000 pages of free embedding credits]
Qdrant

For job seeker:
Find job applications that are best suited for you

For job poster:
Find job seekers that are best suited for your job posting

Normalized score between each resume / job posting pair

[srbhr]

One more would be to add generative capabilities to enhance resume then and there only. If your score it low, then suggest prompts and regenerate certain areas of your resume to enhance it. Co:here also has that. And if we can find a way to integrate an LLM's API then it can become more powerful as a tool.

[srbhr]

Hi, @by14, thanks for the suggestion. Regarding the embedding co:here also provides an API for embedding. For learning there's a playground API that can be used.
Qdrant already has the integration with co:here.

[HunnySajid]

Finalised stack + repo implementation as per today's discussion.

[HunnySajid]

For frontend styling framework, i would suggest to use tailwind + daisyUI for simplicity.

Whichever the framework we decide, I would suggest to consider default components as inspiration while we create designs/mockups. This way design to code implementation would be seamless.

E.g if we go with daisy ui then design should look and feel like its components.

[srbhr]

Roadmap Meeting 2

Things that were discussed:

• Continued UI work by @sandipkarmokar
• Backed by @SubramanyamChalla24
• Fixed single file error by @Hazannovich
• Frontend development in react will start by Next week, after 15th August.
• Some frameworks are being tested by ChickenPasta on Discord. ( @lekhanag246 )