Inspiration
The inspiration came from the common business pain point: the time lag between receiving data and deciding what to do with it. Most tools focus heavily on visualization. I wanted to create a tool that mimicked the result of a high-paid consultant: a concise, strategic report.
What it does
Insight Pilot is a rapid-analysis, high-fidelity web application built to serve as a strategic engine for businesses. The core goal was to bridge the gap between messy raw data (CSV) and immediate, actionable executive strategy. Rather than just visualizing numbers, InsightPilot provides a mock AI-generated analysis—complete with narrative summaries, root cause hypotheses, and concrete action plans—all within a modern, resilient, and polished dashboard interface.
How we built it
The entire project is built using a minimal, yet powerful, stack: Structure & Logic: I used Vanilla JavaScript (ES6+) for all core logic, including CSV parsing, data aggregation, Local Storage management (for history/theme persistence), and DOM manipulation. This kept the application lightweight and fast. Visualization: I integrated Chart.js to create a compelling, dual-axis chart comparing Sales Units against Customer Churn Rate, providing instant visual context for the AI analysis. Design & UX: I implemented a strong CSS Grid layout for the dashboard, ensuring a reliable 75%/25% split for main content and the sticky sidebar. All colors were managed using CSS Variables to enable the instant Dark/Light mode toggle.
Challenges we ran into
Time Crunch on Entry: Joining the hackathon late severely limited our development window, forcing an intense focus on core functionality and polished presentation over deep feature breadth. Actual AI Integration (Scope vs. Time): The original vision included a live API call to a large language model (LLM) for real-time strategic analysis. Due to the limited time, I pivoted to a high-fidelity mock AI engine to prove the UX and value proposition, leaving the final API integration as the immediate next step. Maintaining Visual Stability: Ensuring complex visual features—like the fluid, expanding header and the sticky sidebar—remained responsive and bug-free across desktop and mobile browsers required significant time that had to be carved out of feature development.
Accomplishments that we're proud of
I am immensely proud of two key victories that represent a massive leap in skill, especially given that my web design capabilities were initially very low. First, i achieved a level of visual polish and UX fluidity far beyond our starting point. We mastered a complex, high-fidelity CSS Header Transition, which involved synchronizing the header's height, font size, and the layout (one line to two lines) with smooth animation. This required extensive debugging and refactoring, but the result is a professional, fluid, and bug-free user experience. Second, a robust, resilient dashboard structure, built on a clean CSS Grid with a fixed 75%/25% layout, implemented a complex Sticky Sidebar for premium usability, and ensured the whole application is fully responsive. This foundation proves rapid skill acquisition and leaves me with a clean, high-performance, and easily scalable application.
What we learned
I gained a deeper understanding of synchronizing CSS transitions across parent and child elements, proving that a high-quality, professional UX is achievable with careful Vanilla CSS and JavaScript structure.
What's next for Insight Pilot
The next steps are entirely focused on transitioning InsightPilot from a high-fidelity prototype into a production-ready analytical tool. The immediate priority is replacing the current mock analysis with a direct, live connection to a large language model (LLM) API. This final step will allow the application to generate genuine, context-aware strategic reports by sending the user's sanitized data and a detailed prompt to the service, validating the core functionality of the entire platform. Beyond true AI, we will expand the application's data capabilities by introducing multi-file input support for complex cross-functional analysis and implementing more advanced time-series analysis in our charting features to allow users to compare performance over different fiscal periods. Finally, I will focus on scalability by implementing user authentication for secure, private report history and deploying the application to a robust cloud environment to make the tool stable and globally accessible.
Log in or sign up for Devpost to join the conversation.