AImploy

Links

• https://github.com/HawkHackQwerty
• https://github.com/HawkHackQwerty/Frontend
• https://github.com/HawkHackQwerty/Backend

Inspiration

• When thinking about major issues that students face currently, we found that the job market is fairly averse to those with little experience. It can be quite difficult to break into the industry initially. Moreover, the job market has not been booming very much for internship/full-time roles—due to economic conditions, but also the threat of AI taking away jobs. We wanted to flip this idea on its head. 

• Meet AImploy (a pun on AI + employ)—an all-in-one tool that actually leverages AI to help people land their dream job. 

What It Does

• AImploy is a one-stop shop to assist you on anything related to your job-hunting journey!

• We enable users to seamlessly create cover letters tailored to specific roles and their past experience, receive detailed feedback/a score on their resume, and perform personalized interview preparation through an online recording tool

• PDF upload functionality for resume and cover letter

• Learns from user’s past work experience and current resume to recommend tailored feedback, just for them!

• Gives user interview prompts, records user’s response, and provides constructive feedback and relevant follow-up questions

How We Built It (Tech Stack, Software we Used)

• Software: VSCode, Neovim, Cohere API

• Framework (Frontend): React and Chakra UI

• Framework (Backend): RESTApi using Gin

• Language (Frontend): TypeScript (React) and HTML/CSS

• Language (Backend): Python, GoLang, and Dockerfile

Challenges We Ran Into

• Using shared memory

• Busy study spots

• Working on no sleep

Accomplishments That We're Proud Of

• Creating a cool project!

• Linking the frontend to the backend

• Successfully implementing webpage routing in React

• Connecting the frontend to the backend, leveraging the Cohere API for machine learning and natural language processing

• We effectively collaborated to come up with the idea within the first day, and then assigned each other roles and locked in.

What We Learned

• Choice of Language: During tight time schedules, programmers have to balance out a bunch of factors in their applications, namely, performance and complexity. It is usually said that when one increases the other does as well, but after using GoLang, I can clearly say that isn’t true. Python as well, recently more and more frameworks have been performant since they are written in Rust and Zig but they can compile to C which allows for high performance.

• System Architecture: During the system planning phase, we overestimated our capabilities and had questionable system architecture suggestions like, Using GoLang as our backend and passing it to another python piece using a different connection to be processed. Clearly, this is inefficient, but we were blinded by the desire to learn GoLang during this hackathon.

• How to route pages in React using react-router-dom

• We can down a spicy 2x buldak ramen bowl in an average time of 33 seconds

• Easier is better

What's Next?

• Extension: For the job title and description, they have to manually enter it, but we were planning to make a small extension that does that for them and also acts as an application advisor when on LinkedIn.

• Suggest relevant jobs and relevant employers by searching using RAG for better information about jobs

• Creating a social platform/forum where people can speak to one another for advice—using cryptocurrency/NFTs as incentives.

• Implementing LinkedIn scraping to get hiring managers and job profiles/links

• Integrations with Glassdoor/Indeed/LinkedIn