Game Knowledge Assistant

The Challenge

What was the problem?

Today's gamers face a relentless flood of updates from weekly patch notes and new metas to evolving community builds. Staying competitive means spending hours on Reddit, YouTube, or patch sites. This overwhelms even the most dedicated players.

Key pain points we identified:

• Difficult to keep track of what's changed in your favorite games.
• Forums & wikis are scattered, long, and often outdated.
• Players feel left behind when the meta shifts.

My Solution

The idea

I designed an AI-powered gaming assistant that:

• Summarises patch notes & community findings into bite-sized insights.
• Answers strategy & meta questions instantly: “How do I counter the new Zed build?”
• Lets players track only the games they care about, delivered in their favorite platforms like Discord & Twitch.

The payoff

It's like having a personal esports analyst, 24/7.

Research

To start, I investigated how players across communities like League of Legends, Valorant and World of Warcraft stay updated. Most still rely on lengthy patch notes, YouTube explainers or Reddit discussions, which are often fragmented and time-consuming. I also used ChatGPT to research which game titles would be most compelling to include. Alongside this, I performed a competitive analysis of tools such as Mobalytics and Tracker.gg, which revealed that while these platforms offer deep insights, none provide personalised, multi-game AI summaries that work seamlessly across Discord, Twitch and mobile. This highlighted a clear opportunity in a very niche market.

Player personas

Using ChatGPT, I developed two gaming personas based on the research findings. The first was the Competitive Climber, a 23-year-old League player pushing for Diamond who wanted fast, direct insights to stay ahead of the meta. The second was the Social Gamer, a 27-year-old who plays casually with friends and values quick, friendly summaries without heavy detail. These personas kept the project anchored in authentic player needs and shaped every design and product decision.

Site map

I used Microsoft Copilot to generate a sitemap outlining the core areas of the experience. This included a dashboard with personalised summaries of patch notes, trending builds and upcoming tournaments, along with a chat assistant that could answer live questions inside Discord or Twitch. It also featured settings where players could pick the games they wanted updates on to avoid notification overload. I later expanded this sitemap to include dedicated games pages, individual game detail views and a user profile page to track preferences and history.

Concepts

With a clear direction, I used ChatGPT to draft a comprehensive product prompt which I imported into Google Stitch in experimental mode. This helped generate a range of early concepts and basic wireframes that I could then take into Figma for refinement. Through this process, I focused on creating a mobile app and website, giving players flexibility to stay informed wherever they were. User testing reinforced that there was strong demand for a product like this, confirming we were addressing a real gap in the market.

Wireframes

I imported the initial outputs from Google Stitch into Figma, then adapted and expanded on them by creating new elements and refining flows such as onboarding, the chat Q&A experience and the multi-game dashboard. In some cases, visual elements didn't import cleanly, so I corrected and rebuilt them directly in Figma. I also continued to use Stitch to explore additional ideas, which were brought back into Figma to accelerate the design exploration and ensure a robust foundation.

Visual design

For the visual style, I drew inspiration from esports broadcasts and gaming dashboards, using dark themes with neon accents to capture the energy of competitive play. I prioritised high contrast for easy readability and designed modular card components to keep information clear and scannable. This visual system was built to scale across mobile, web and future Discord or Twitch integrations, maintaining a consistent brand presence everywhere.

Prototyping

Using Figma's AI features, I created interactive prototypes that simulated asking the AI questions and receiving concise, bullet-point responses. I also built flows for the dashboard where players could toggle between games like Valorant and League. This helped validate the core experience, and small enhancements such as adding emojis to highlight key updates noticeably increased engagement in early tests.

Build

To bring everything together, I used the detailed product prompt created in ChatGPT and copied it directly into Firebase Studio, alongside the Figma designs, to build a fully working product. This ensured the app was both functional and visually consistent. I also explored Lovable to test alternative AI-driven build workflows, giving me extra flexibility. The result was a fully responsive product that worked beautifully across mobile and web.

Outcome

The final concept was an AI-powered gaming companion that delivered exactly what players needed, wherever they spent their time. Beta testing highlighted a clear demand for a product like this, especially in a space that's still highly niche and underserved. The project showed how AI can personalise complex gaming data, and how thoughtful UX design can transform overwhelming information into a true competitive edge.