This is typical portfolio page with a creative twist. The portfolio page also has an Assistant AI chat bot named Vada Pav which can answer questions for the person who's portfolio is being showcased. He can answer questions as if it were an interview. This helps the interviewer to know what he wants through the AI and skip any unwanted data shown in a regular portfolio.
What tools did you use to create your project?
How much experience does your group have? Does the project use anything (art, music, starter kits) you didn't create?
Our group has a solid foundation in programming, especially in Python, AI/ML concepts, and web development. We’ve worked on several small to mid-scale projects before, both individually and as a team. For this project, we used pre-trained models from Ollama, and while the core logic and data integration were custom-built, we did not create the AI models ourselves. No external art, music, or starter kits were used — everything else was built from scratch.
What challenges did you encounter?
Some key challenges we faced:
Integrating LLMs locally using Ollama and ensuring they responded accurately to portfolio queries.
Data structuring, so that the AI could understand and provide relevant answers.
Keeping the conversation natural, without the bot hallucinating or giving unrelated information.
Making the interface user-friendly, especially for those who might not be tech-savvy.
And, of course, optimizing performance since local models can be resource-heavy.