Decision OS

Inspiration

Decision OS was inspired by a simple problem: people often make important decisions with incomplete structure. They know their goal, but the options, constraints, risks, and tradeoffs are usually scattered in their head. We wanted to build a tool that turns a messy natural-language decision into a clear, reasoned analysis.

What it does

Decision OS helps users analyze decisions by entering their goal, options, and relevant facts in plain English. The app identifies the core decision, extracts the goal and constraints, evaluates each option across factors like goal fit, feasibility, risk, and time impact, then provides a recommendation with a concise decision insight and action plan.

How we built it

We built Decision OS as a structured decision-analysis web app. The frontend focuses on a clean input-and-results flow, where users describe their decision and receive organized cards for each option. The backend logic parses the user’s decision, identifies options and constraints, scores each option, and generates an explainable recommendation instead of just giving a generic answer.

Challenges we ran into

One of the biggest challenges was avoiding vague or generic advice. Early versions could recommend obvious steps like “clarify the decision,” which was not useful when the user had already provided enough information. We had to improve the analysis rules so the app focuses on the actual options, tradeoffs, and goal alignment. Another challenge was making the recommendation feel specific while still being flexible for different types of decisions.

Accomplishments that we're proud of

One of the biggest challenges was avoiding vague or generic advice. Early versions could recommend obvious steps like “clarify the decision,” which was not useful when the user had already provided enough information. We had to improve the analysis rules so the app focuses on the actual options, tradeoffs, and goal alignment. Another challenge was making the recommendation feel specific while still being flexible for different types of decisions.

What we learned

We learned that decision-making is not just about picking the “best” option. A good decision tool needs to consider context, constraints, risk, time cost, and the user’s actual goal. We also learned that explainability matters: users are more likely to trust a recommendation when they can see the reasoning behind it.

What's next for Decision OS

Next, we want to add user accounts, saved decision history, better comparison charts, and more advanced scoring customization. We also want to improve the analysis engine so users can adjust priorities, add new constraints, and revisit past decisions to see whether the recommendation worked in practice.

Built With

• medo
• typescript