.md

Examples

Real-world examples of what you can ask the Race Engineer to do.

You don’t need to write any code yourself. Just ask your AI assistant in natural language — it writes the TypeScript behind the scenes.

Coaching & Analysis

”Where am I losing time at Spa?”

The AI will:

  1. Find your recent sessions at Spa
  2. Run getWeakCorners to identify problem areas
  3. Pull braking zone data for your latest session
  4. Generate a coaching report with specific improvement actions

”Compare my last two laps and tell me what changed”

The AI will:

  1. Get your most recent session with getSessions
  2. Find the last two valid laps via getSessionDetail
  3. Run compareLaps for zone-by-zone braking deltas
  4. Explain the differences in trail braking, entry speed, and exit speed

”Am I getting faster at Monza?”

The AI will:

  1. Call getProgressOverTime with trackName: "Monza"
  2. Analyze the trend direction and slope
  3. Compare best/average lap times across sessions
  4. Report whether you’re improving, plateauing, or regressing

Sample-Level Duels

Zoom into the raw telemetry to compare the shape of your driving — sample by sample — against your own laps or the leaderboard leader. (Requires a Plus or Ultra plan.)

”Duel my best Spa lap against the leaderboard leader”

The AI will:

  1. Find your best Spa lap via getSessions + getSessionDetail
  2. Pull the current leader for the same car and track from getDesktopLeaderboard (each entry carries a lap ID)
  3. Run compareTraces with isPublicB: true to align both laps on a distance grid
  4. Report exactly where — and by how much — you gain or lose time, with the cumulative delta at every point

”Where am I understeering or oversteering?”

The AI will:

  1. Take your best valid lap for the session
  2. Run getDrivingSymptoms to diagnose each corner
  3. Flag understeer, oversteer, and traction limits — each with the numeric evidence behind it
  4. Suggest whether it looks like a driving or a setup fix

”Which braking zone am I least consistent in?”

The AI will:

  1. Run getZoneConsistency across every valid lap in the session
  2. Rank the zones by how much your brake point, pressure, and min speed vary
  3. Pull the sample-level trace of the worst zone with getZoneTrace
  4. Show you the repeatable brake shape to aim for

Race Preparation

”What races are coming up this week in iRacing?”

The AI will:

  1. Call searchRacesThisWeek to get all active series
  2. Filter by your preferred category (road, oval, etc.)
  3. Show tracks, times, and license requirements

”Create a race event for the IMSA Pilot Challenge at Watkins Glen this Saturday”

The AI will:

  1. Search searchIRacingSeries for the IMSA series
  2. Find the current week’s schedule with getSeriesSchedule
  3. Call createRaceFromSchedule with the matching series and week
  4. Confirm the race was added to your calendar

”Generate a fuel strategy for my 45-minute race”

This example needs an Ultra plan.

The AI will:

  1. Find your practice sessions at the race track
  2. Call generateStrategyFromTelemetry with your session data
  3. Calculate fuel per lap, tank range, optimal stint length
  4. Present a stint plan with pit windows and fuel loads

Track Notes

”Update my Spa notes with what I learned today”

The AI will:

  1. Find your existing Spa notepad with getTrackNotepads
  2. Read the current notes with getTrackNotepad
  3. Merge your new observations into the corner notes
  4. Save with updateTrackNotepad

”What did I write about Turn 1 at Barcelona?”

The AI will:

  1. Search your notepads for Barcelona
  2. Find the corner note for Turn 1
  3. Return your braking reference, gear, min speed, and personal notes

Advanced Analysis

”Give me a full readiness assessment for Saturday’s race”

The AI will:

  1. Call getRaceReadinessData with your race event
  2. Analyze all practice sessions (FP1, FP2, FP3)
  3. Evaluate pace, consistency, phase performance, and corner evolution
  4. Save a saveRaceReadinessReport with strengths, weaknesses, and race plan

”Which car/track combos have I driven the most?”

The AI will:

  1. Call getSessions with a high limit
  2. Group sessions by track and car combination
  3. Calculate total laps, hours, and best times per combo
  4. Present a ranked summary

Setup Decisions

(Setups is in Public Beta and requires your own Plus or Ultra personal plan. A Team seat alone does not grant access.)

”Would one click toward less negative rear camber make sense here?”

The AI will:

  1. Resolve the exact setup, version and compatible session you mean
  2. Read the current adjustable values, telemetry context, your confirmed experiment history and reviewed setup knowledge when available
  3. Answer the camber question first, label its confidence and state what evidence is missing — it will not call camber optimal without inner/middle/outer tyre temperatures
  4. Offer at most one verified adjustable alternative, with only one or two settings in the proposed test
  5. Leave the setup untouched

If the direction makes sense to you, follow with an explicit request:

“Apply that one-click rear anti-roll-bar test as a new version.”

The engineer uses physical directions — “1 click softer” or “1 click on each rear side” — rather than internal indices. It only shows current → proposed values when that garage mapping is reliable.

Tips for Better Results

  • Be specific: “Analyze my braking at Turn 3” works better than “help me improve”
  • Mention the track: Helps the AI filter to relevant data quickly
  • Ask for reports: The AI will save coaching reports visible in the web app
  • Chain requests: “Find my weakest corner, then duel that corner against the leaderboard leader”