apache / apache/texera

Add Bet Pilot dashboard prototype

Open
#5,109 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Scala
Stars
314
Forks
187
Avg merge
1d 21h
Merged PRs (30d)
214

Description

### Feature Summary

Bet Pilot is a hackathon prototype that shows how Texera can host a domain-specific decision app on top of imported workflows. It gives users a dashboard for Valorant prop-bet analysis with daily recommendations, manual line entry, model health, bankroll, calibration, and workflow preview screens.

### Proposed Solution or Design

Add a new authenticated dashboard app entry under `Dashboard > Apps > Bet Pilot`.

| Area | Proposed behavior |
| --- | --- |
| Daily picks | Summarize selected and skipped lines with clear model reasoning. |
| Lines input | Let users paste manual lines or stage image filenames for the workflow pipeline. |
| Scouting | Explain one candidate pick with recent match history, map scenarios, residual factors, and confidence. |
| Model health | Show calibration, coverage, and edge slices from stand-in data. |
| Bankroll and calibration | Display risk sizing and model refit history. |
| Workflow previews | Link dashboard screens back to imported Texera workflow ids and render small DAG previews. |

This is related to the dkNet-AI hackathon idea discussion: https://github.com/apache/texera/discussions/5059

### Affected Area

- Workflow UI
- Other

Contributor guide

Open the contributing guide

Research direction

Start at the authenticated Dashboard > Apps entry point and review the related dkNet-AI hackathon discussion before scoping the prototype. Done should include the listed daily picks, line input, scouting, model health, bankroll, calibration, and workflow preview areas, with screens linked to imported workflow IDs.

Written by the indexing model from the issue text.

Assessment

Domain
frontend, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
28/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.