Team63_6F_TravelBuddy

TravelBuddy is your personal AI-powered travel companion that revolutionises travel. It is your gateway to exclusive partnerships and experiences. Join us in redefining travelling.

Comment

Team Number: 63 Reference Code: 6F

Inspiration

With the recovery of the tourism sector from the Covid-19 pandemic, there is a growing need for innovative solutions to address the changes in travel behaviour and ensure a seamless and enjoyable experience for travellers. Users tend to be overwhelmed with existing travel planner applications such as Wanderlog or Roam Around. They struggle to sift through vast amounts of suggestions, including travel guides, reviews, and real-time updates, to distil the ideal recommendations. As such, we seek to leverage the strengths of those applications while improving on their current limitations to provide the best travel experience for all our users.

What it does

By simply inputting your trip details and preferences, TravelBuddy's advanced AI chatbot engages you in natural conversations, crafting a tailor-made itinerary that not only meets your accommodation and dietary preferences but also introduces you to exclusive experiences and hidden gems. Our unique algorithm adapts to your choices in real time, ensuring each trip is as unique as you are.

Beyond planning, we seek to include gamification features in our application by rewarding user contributions to our community with loyalty points redeemable for vouchers and discounts. Not only will this improve our recommendations, it will also foster a network of passionate travellers. There are various marketing and partnership opportunities on our application as our chatbot seeks to inform and promote services and activities that are suitable to our users’ tastes and preferences.

How we built it

We decided to use the popular t3 stack, popularised by the YouTuber theo.gg. The T3 Stack combines TypeScript's type safety, Tailwind CSS's utility-first styling, and tRPC's end-to-end typesafe APIs for a streamlined development process. It enhances code reliability, speeds up design implementation, simplifies API handling, and ensures scalability, offering a cohesive, efficient, and modern approach to web development. We used OpenAI's GPT-4 turbo model, which is more powerful than ChatGPT Pro. For data storage, we used s3 buckets for images and a Postgres database created within a docker container. To interface with our database, we made use of the drizzle-orm. We used the Google Places API to get relevant photos of locations in our itinerary, as well as, clerk library for authentication. We felt that this was important as they are renowned for securely storing user credentials. Lastly, we used MaterialUI, as it has pre-built React components, our frontend webpages.

Challenges we ran into

Throughout the hackathon, our team quickly recognised the critical role of effective communication. Repeatedly, we encountered moments where our perceptions of a particular application feature diverged, prompting in-depth discussions and repeated clarifications to achieve alignment. This experience taught us the value of ensuring mutual understanding such as through creating mockups or wireframes before proceeding.

Additionally, we faced challenges with git, a tool unfamiliar to us as first-year students. The initial learning curve was steep, but by the end of the hackathon, our proficiency had noticeably improved. This newfound comfort with git is something we eagerly anticipate applying to future projects.

Accomplishments that we're proud of

We take great pride in the AI chatbot we created, which exceeded our expectations with its precision and ability to engage users with contextually relevant questions, thus crafting optimal travel itineraries. Its intuitive interaction and intelligent responses significantly enhanced the overall user experience. Additionally, we are equally proud of the website and its features. It effectively addresses the challenges we aimed to tackle, disrupting the market by offering a significant value addition to users and potential partners. The website's user-friendly interface and seamless navigation further reflect our commitment to providing a solution that is not only functional but also aesthetically pleasing and easy to use.

What we learned

Throughout the hackathon, our team embarked on a steep learning journey, embracing the nuances of effective team communication and the intricacies of git. These experiences not only enriched our technical capabilities but also deepened our understanding of collaborative innovation in a high-pressure environment.

More significantly, we learnt that it is essential not to simply replicate or merely improve an existing product or service available in the market. It is important to brainstorm and think of innovative ways to differentiate our product/service and punch a hole in the industry. Aside from the core practicalities of the application, we learnt that it is also necessary to incorporate business elements to make our application sustainable and viable in the long-term. Overall, we understood better about the need to find the delicate balance between innovation and practicality, especially in the capitalist society that we live in today.

What's next for TravelBuddy

Future options are endless for TravelBuddy. With improvements in databases and machine learning, our chatbot would be able to differentiate trips more distinctly for each unique individual. Following various up and coming travel trends, TravelBuddy will also be able to accurately configure trips specially catered to non-conventional travel destinations. For instance, TravelBuddy may include options that are more eco-sustainable to promote green tourism. TravelBuddy may also have the option to create road trips or hiking trips which have been deemed to be more challenging to plan.

In terms of partnerships and profitability, our chatbot can also be more proactive in recommending different types of services. This may include accommodations, flights or even activities commonly found on agency sites such as Klook. We may also consider implementing a tiered subscription service to offer more premium recommendations and services to our regular subscribers. With the partnerships forged and funds received, we can better improve our algorithms and features, bringing TravelBuddy to even greater heights!

Built With

Share this project:

Updates