Orna

Outdated career planning fails today's dynamic job market. Our AI-driven tool, using ML & psychology algorithms, maps careers with LLM and RAG insights offering students tailored, strategic blueprints

Comment

Inspiration

Selin's journey was the spark that ignited the creation of our platform. Originally diving into the world of chemistry, she believed it was her calling. However, as time unfolded, she realized it wasn't the path that resonated with her true passions. This realization, while enlightening, also brought with it a wave of confusion and stress. The weight of expectations, both self-imposed and from the university, pressed down on her, urging her to find a new direction swiftly. Yet, the vast expanse of potential careers felt overwhelming, leaving her adrift in a sea of options, not knowing which shore to swim towards. Selin's story isn't unique. It's a narrative that echoes across university halls, with countless students grappling with the same feelings of uncertainty and pressure. Recognizing this widespread challenge became the cornerstone of our mission: to illuminate the myriad of career paths available and guide students towards their true calling.

What it does

Our platform is an AI-powered career mapping tool designed for students navigating the tech landscape. Utilizing advanced machine learning algorithms combined with psychology-driven techniques, it offers a dynamic and data-driven representation of potential career paths. Each job node within the map is informed by LLM and RAG methodologies, providing a comprehensive view based on real user trajectories and data. Beyond mere visualization, the platform breaks down tasks into detailed timelines, ensuring clarity at every step. By integrating insights from both AI and psychology algorithms, we aim to provide students with a clear, strategic blueprint for their ideal tech career.

How we built it

We integrated advanced machine learning algorithms with psychology-driven techniques. The platform's backbone is built on LLM and RAG methodologies, informed by real user trajectories. We also incorporated various APIs, like the Hume AI API, to enhance user experience and data collection.

Challenges we ran into

Embarking on this journey, we were rookies in the arena of hackathons, stepping into uncharted territory with a blend of enthusiasm and trepidation. The path was riddled with unexpected hurdles, the most formidable being a persistent bug in the RAG model from MindsDB. Hours that could have been spent refining and enhancing were instead consumed in troubleshooting this elusive issue. As if the technical challenges weren't daunting enough, the spirit of the hackathon was challenged as friends, one after another, decided to step away despite having amazing ideas. The weight of their absence, combined with the mounting pressure of having to reconstruct a new model (the bug turned out to be withing MindDB's RAG handler which we had no control over) in a race against time, was palpable. With the clock ticking, sleep became a luxury we could scarcely afford, operating on a mere three hours. Yet, in the face of these adversities, it was our shared vision and unwavering determination that became our beacon, guiding us through the darkest hours and reminding us of the potential impact of our creation. The true essence of our challenge wasn't just in navigating technical glitches or decreasing excitement in the event; it was about resilience, adaptability, and the relentless pursuit of innovation.

Accomplishments that we're proud of

Successfully integrating the Hume AI API to translate user opinions into actionable data was a significant win and as well as connecting it into a spreadsheet for further analysis. Despite the hurdles and it being our inaugural hackathon, our team's perseverance saw us through to the end.

What we learned

We gained a lot of insignts into the nature of LLM models and the intricacies of the RAG model. The experience also taught us the importance of adaptability and persistence in the face of unforeseen challenges.

What's next for Orna

Our immediate goal is to finalize the MVP, refining the suggestion system. With these enhancements, we aim to secure seed investment to propel Orna to new heights.

Share this project:

Updates