Keep the evidence chain inspectable
Summaries should be traceable back to the events that produced them instead of appearing as unexplained totals.
Real-time game data tool
Neverwinter Live Parser
A technical side project about transforming raw events into useful evidence without hiding data quality.
- Role
- Product Designer + Front-End Builder
- Year
- 2025
- Product
- Real-time game data tool
- Users
- Neverwinter players using combat data to understand encounters and performance
Raw events are plentiful. Trustworthy information is harder.
A live combat parser has to ingest repeated events, group them correctly, calculate useful summaries and communicate when the underlying data is incomplete or ambiguous. The interface therefore had two jobs: make dense performance data scannable and make the limits of that data visible enough that users do not over-trust a clean number.
My contribution. I worked on the information model, parser-facing UI, aggregation logic presentation and the relationship between raw events, calculated summaries and confidence. The project also let me work directly with implementation constraints rather than handing the design to another discipline.
Where the design judgment mattered.
Separate signal from noise
The interface prioritizes the comparisons and changes that affect a decision rather than rendering every available value with equal weight.
Show uncertainty instead of polishing it away
Missing, delayed or ambiguous data is treated as a state the interface must explain.
Screens, shipped material and public references.
The artifacts sit inside the story because proof is more useful next to the decision it supports than in a ceremonial section at the end.
This case has limited public visual material, so the write-up stays focused on the work I can describe accurately rather than filling the page with decorative substitutes.
Experience model
From event stream to decision
The project demonstrates the same design concern as a business dashboard: information is only useful when people understand where it came from, what it means and how confident they should be in it.
- 01
Capture raw combat events.
- 02
Normalize and group events into useful entities.
- 03
Aggregate metrics with explicit assumptions.
- 04
Present the result with enough context to judge confidence.
What this work demonstrates
A data tool that makes its assumptions visible
The project demonstrates the same design concern as a business dashboard: information is only useful when people understand where it came from, what it means and how confident they should be in it.
Data-product thinkingTechnical implementationConfidence and uncertainty UX