Squidfort

• 

  Landing page (image courtesy of DALL•E)

• 

  Profile form

• 

  Results page (unfinished!)

Inspiration

As a freelancer, finding work can be time-consuming and exhausting. There are many marketplaces to discover gigs, but most of us would rather spend time doing what we're hired for than scouring the internet for prospects. With the recent boom in language models making outsourced intelligence so accessible, it made sense to build something that tackled this problem. The changing world of work has me convinced that personalised AI will become the companion advising us throughout our careers, freeing us from busywork along the way.

What it does

The app takes your professional profile (currently allows LinkedIn data imports) and recommends gigs based on your work experience, education, location, and interests.

How we built it

I built the app using React, TypeScript, LangChain and Weaviate amongst other tools.

Challenges we ran into

I ran out of time!

Accomplishments that we're proud of

Escaping perfectionist traps (mostly).

What we learned

There's a lot I don't know about RAG, though I believe the field is still evolving. It gives me more material to read up on!

What's next for Squidfort

I'm planning to expand the project dataset to Upwork projects and a viable bounty platform too.

Built With

• claude
• langchain
• openai
• react
• typescript
• weaviate