HackStack

Project Info

flowzone

Devpost

Inspiration

Our team is composed of aspiring and existing entrepreneurs, where we share a common struggle of having multiple goals but insufficient time to achieve them all. Goal optimization then stood out as a problem we wanted to tackle. Likewise, research shows that 92% of people fail to achieve their goals. After doing extensive market research, we realized there are productivity tools that aid goal optimization like Todoist which breaks down goals into actionables, or Motion which schedules actionables into your calendar. While current solutions work in silos, we see value in having a comprehensive end-to-end solution from goal setting to Simplification to Integration and Analysis. This birthed flowzone. With flowzone, you can halve the time you spend achieving your goals. A one-year goal shortens to merely 6 months.

What it does

From setting your goals to simplifying them into actionable tasks, to integrating them into your existing schedule, to analyzing your progress - flowzone is there every step of the way. flowzone is your end-to-end solution and best pal to help you achieve your goals. Simplify - Our AI breaks down your goal into a list of actionable tasks, with quantifiable time frames, segmented into phases. Integrate - Our AI analyzes your existing calendar (eg. synced Google calendar), and slots in your actionables on top of your existing routines. Optimize - Our AI analyzes both quantitative data (eg. stress levels and sleep patterns from TerraAPI), as well as qualitative data (from our daily reflections form where users reflect on their wins and losses), to recommend better ways to optimize your path to achieving your goals. flowzone knows you like the back of our palms - we anticipate your needs and wants before you realize them yourself. Knowing that having poor sleep quality drops your focus time to 2 hours at a time, we will suggest shortening your work intervals to suit your productivity trends. Knowing that your peak productivity is from 8am-10am, we will suggest scheduling High Priority action items during those hours. The list goes on.

How we built it

Next.js for front-end Convex for back-end TerraAPI to get health data from wearables TogetherAI for AI agent and analysis

Challenges we ran into

Training and finetuning our models in a short period of time - there is still much we would like to work on to achieve close to 100% accuracy. Improving inference speed was also a major challenge we faced. Integrating all the different APIs and linking the various databases into a cohesive system and flow for our users was also something we had to try and retry to find the most intuitive user flow, after rounds of user testing.

Accomplishments we're proud of

Being able to dissect the issue of goal optimization into concrete features addressing the 3 key pain points in a short 2 days. Due to the complexity of the solution we wanted flowzone to be - with the multiple integrations of API, linkages in database, extensive model training, we all had to step out of our comfort zone to fulfill the needs of the team. For example, our front-end developers had to double up as back-end, our back-end developers had to double up as AI developers, etc.

What we learned

We learnt the importance of adopting the iterative design thinking process. As we developed, we made sure to use an iterative feedback loop - doing regular user testing to ensure we are meeting their pain points and designing in an intuitive manner. This is also an idea we really believed in so trying to build a product that would make sense as a business, and taking into consideration our target market, business model, etc also held weight on our design of the product.

What's next

Short-term milestones: Mar 2024 - MVP Jun 2024 - Beta Launch Sep 2024 - Live Mobile App Next phase milestones: Explore Enterprise plans for teams on top of current Freemium models for individuals. Further finetune AI models to provide scientifically / academia-proven recommendations and analysis.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • CSSIn code
  • ExpressIn code
  • JavaScriptIn code
  • MongoDBIn code
  • Next.jsIn code
  • OpenAIIn code
  • ReactIn code
  • Tailwind CSSIn code
  • TypeScriptIn code

9 of 9 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

276 KB

Source files

105

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars