Switching careers is a hurdle: can AI ease the transition with personalized learning paths?
HackYeah is the biggest on-site hackathon in Europe. I teamed up with four developers in the 2024 edition to take part in the education category challenge. In just 24 hours, we developed a functional solution, securing 6th place out of 62 teams.
get_hired demo
Project background
Imagine this: You’re feeling stuck in your career, and thoughts about changing your professional path start to pop into your head. You know you have skills, but you’re unsure how to position yourself, what to focus on, or even what direction to take.
Our solution, get_hired, is like a personal career assistant that helps you understand your current skills and guides you step-by-step on how to get to where you want to be. With personalized assessments and tailored learning paths, it ensures that each user receives a unique experience aligned with their skills and aspirations.
As the solo designer on our team, my task was to bring the concept to life visually. I mapped out the user flow, created wireframes and high-fidelity designs, and put together the presentation.
What made our project stand out from the jury's perspective?
It was our first time at HackYeah, and to be honest, we didn’t have high hopes, given the number of participants. Tauron Arena was packed to the brim. Nevertheless, we enthusiastically started brainstorming ideas as we wanted to make the most of the experience. After 24 hours of hard work, we were quite happy with the result. Unfortunately, we didn’t make it to the finals, but…
We asked the mentors who reviewed our project for feedback, and it turned out that we were just 0.1 points short of reaching the finals, achieving a score significantly higher than average. The jury highly praised the idea, design, and presentation.
How the idea for get_hired was born
As we brainstormed, we reflected on our own experiences and the challenges we faced on our paths to new careers. We realized that many people struggle with navigating professional transitions. Facebook groups, Quora, and Reddit are filled with questions from individuals looking for guidance on how to switch careers, where to start, and what to learn. It can feel overwhelming and confusing.
Also, everyone’s journey is different, so cookie-cutter bootcamps or courses don't always cut it. What is more, they can be really expensive while offering disappointing value.
So how about creating a platform that adapts to the learner? With machine learning and AI it's quite possible. Actually, this is how the future of self-directed learning may soon look like.
How it works? Mapping out user flow
This was our base scenario: Ela, a marketing specialist, feels burnt out and wants to change careers to a field where she can use her data analysis skills and creativity in problem-solving. She's considering UX design but isn't sure what skills she's missing or where to start. She discovers get_hired through an ad. She is curious to try it out.
- Ela signs up to get_hired via Gmail.
- She lands on a dashboard.
- She clicks "get started" button, which takes her to a chat with an AI career coach.
- She answers a few questions about her experience, strengths, and career aspirations.
- She gets a personalized development plan, tailored to her profile. She can access it by clicking "Go to my course".
- She lands on a course page, where she can see all the lessons and track her progress.
- She clicks "Start course" to begin her first lesson.
Wireframing and designing
The clock was ticking, so I had to prioritize clarity and functionality over a fancy design. I quickly created very simple wireframes, so frontent developers could start working on their part, and then I focused on the visual layer.
To speed things up, we decided to use shadcn UI library. We also agreed that I would first create the "minimum" design, which would serve as a base ready for implementation. In the meantime, I would continue developing it, and if time allowed, the changes would be introduced into production. Unfortunately, technical difficulties cost the developers some time, and they didn't manage to implement the second iteration.
For the landing page and chat view, we based on a design created during a previous hackathon. This allowed us to implement these more quickly, saving extra hours.
Since there was no time for a thorough design review, there are some differences between the design and the production.
Monetization opportunities
get_hired is meant to work in a SaaS model with 3 main monetization strategies:
Subscription fee
Monthly plan or annual subscription fee to access the platform. Free trail gives access to one course.
Premium feature - 1:1 Mentorship
The 1:1 mentorship feature connects users seeking career transitions with experienced mentors in their desired field. This provides users with personalized feedback from experienced industry insiders.
Premium feature - visibility for employers
This feature allows users to make their profiles visible to employers looking to fill open positions. Employers can browse profiles, evaluate candidates based on skills, education, and course completion, and reach out directly to chosen users.
Technical limiations & what could have been done better
Reflecting on the UX side of the project, I would improve the login process to emphasize that creating an account starts a 7-day trial period. This is an important element of the user experience that I overlooked.
I would also invest more time in rethinking and redesigning the empty state, as it is the first interaction users have with the app. Perhaps this is where the chat with the AI assistant should begin? Or I could consider adding an engaging onboarding tutorial? Testing the app with users, could provide valuable insights.
The biggest challenge of this project, from a technical perspective, is that in order to prepare valuable learning experiences, the AI must be trained on high-quality, expert-curated data, which takes time and resources. Our solution is not yet capable of creating truly valuable course content. Nonetheless, it’s at a good starting point. Also, the chat with AI coach needs improvement to ask more in-depth follow-up questions and lead a more natural conversation.
What this project taught me
Flexibility in working with developers
An iterative approach to design allowed us to accelerate the development stage. By concentrating on a minimum viable design, I efficiently prototyped a functional product, even if it lacked visual polish.
MVP requires compromises
Creating an MVP, within a tight 24-hour timeframe, required a laser focus on core functionalities. Keeping the design simple and using an existing component library sped up development, even if the final product wasn’t as visually polished as I had dreamed of :)
Effective presentation
This already proved very important during our previous hackathon, and was once again highly praised by the judges. A strong rationale behind the idea and product functionality with a pinch of business perspective, is the secret to a successful product pitch.