What is a Track Engineer in sim racing?

What a race engineer does in real motorsport

In an F1, GT3 or any professional racing team, the race engineer is the person sitting on the pit wall, with several screens in front of them, running the car from a technical standpoint. They don't drive. They don't physically touch the car. But every meaningful decision about the car goes through them.

Their job boils down to four things: defining the setup (the car's mechanical and aerodynamic configuration), planning the race strategy (fuel, stops, tyres, safety car management), interpreting the telemetry coming off the car, and communicating with the driver over the radio to make real-time decisions.

It's no accident that they're the closest person to the driver in the paddock. Lewis Hamilton had Peter Bonnington (Bono); Verstappen has Gianpiero Lambiase (GP). The trust built over years in that relationship is worth more than a lot of carbon fibre.

How that translates to sim racing

Here's the interesting part about sim racing: the simulator gives you the same kind of information a real car does, sometimes more. Detailed telemetry, pressures, temperatures, slip angles, G-forces, brake maps. All of that data serves the exact same purpose as it does on a real track — understanding what the car is doing and deciding how to improve it.

The problem is that the sim racer is also their own engineer, mechanic and strategist. And unless you have hours to dedicate to setup work and the experience to interpret the data, that's a massive bottleneck. Most of the time you lose in an online race doesn't come from your driving — it comes from running a mediocre setup or a wrong strategy.

That's why a whole category of tools has emerged in recent years to take over the technical role: strategy planners, telemetry readers, and more recently AI-driven setup advisors. The job a human engineer used to do inside a team is now available for a single driver.

The 4 areas a race engineer covers

To understand what a race engineer — human or virtual — actually gives you, it helps to break the job into its four responsibilities. Each has its own complexity and its own impact on the final result.

All four are connected. A bad setup destroys the telemetry. A bad strategy ruins the best setup. And without clear communication, the driver has no idea what to do differently.

Setup: the most visible area

Setup is what most sim racers picture when they hear "engineer". It's that menu with dozens of values — springs, bars, pressures, geometry, wings, brake maps — that determines how the car behaves.

A good setup isn't one that's "fast" in the abstract. It's one that fits your driving. If you brake late and turn in hard, you need a more responsive car on entry. If you wait for the apex and roll on power early, you need stability on exit. That's why an alien's setup off the internet is usually awful for an average driver: it's built on reference points you don't have.

A well-designed AI — like Track Engineer from ApexSimApps — reads your actual telemetry (not someone else's) and infers what kind of driver you are before suggesting changes. The difference between that and a generic setup is the difference between a tailor who measures you and one who hands you a size M and hopes for the best.

Strategy: fuel, stops, rain

Strategy is the other half of the engineer's job, and the most underrated one in amateur sim racing. Most drivers load fuel by eye, count stops in their head, and react to rain when they're already aquaplaning. That works in short sprints. In endurance, it doesn't.

A properly calculated strategy accounts for:

• Per-lap fuel consumption tuned to your actual pace, not the theoretical one.
• Fuel mass effect on lap time (every 10 kg costs 0.3 to 0.5 seconds, depending on the car).
• Pit lane time loss — how much each stop actually costs you.
• Tyre degradation based on starting pressure, ambient temperature and driving style.
• Rain scenarios — what to do if it starts on lap 12 vs lap 18.

Working all of that out by hand is slow and error-prone. That's why dedicated tools exist, like SimRacing Planner PRO, that automate the math and let you compare strategies before the lights go out. The real-world equivalent is what every F1 strategist does before the start: Monte Carlo simulations over hundreds of scenarios.

Telemetry: reading the car's data

Telemetry is the race engineer's native language. It's what separates an opinion ("the car understeers in the middle of the corner") from a data point ("front axle slip angle goes from 6° to 9° between apex and exit"). One sharpens a feeling; the other tells you what to change.

In Assetto Corsa, the telemetry the simulator generates is professional-grade. You get pressures, three-zone temperatures (inner, middle, outer), wheel speeds, slip angles, G-forces, sector times, exact car position on track. What's missing isn't data — it's someone who knows what to look for in it.

A new driver who opens a telemetry channel stares at 40 graphs without knowing where to begin. An engineer — human or AI — knows there are 3 or 4 patterns that matter in any given case, spots them in seconds and connects each one to a concrete setup change.

To go deeper on reading specific car data, check our Assetto Corsa setup guide, which covers each parameter in turn.

Communication: how you talk to your engineer

Communication is the most human part of the job and the part that breaks the most when you put an AI in the middle. A driver usually describes feelings, not data: "the car feels weird", "I can't get it turned in", "the rear is loose". A good engineer turns those phrases into technical hypotheses and asks the questions that sharpen the diagnosis.

Current AIs — and the one inside Track Engineer in particular — accept exactly this kind of natural-language feedback. You can tell it "the car snaps on exit of Parabolica when I get on the throttle" and it combines that sentence with your telemetry to propose a fix. It's not perfect, but the difference between that and a driver working alone is huge.

What you do lose compared to a human engineer is the emotional layer — knowing when to push the driver and when to back off. But that only matters in top-level competition.

Can AI replace a human engineer?

For amateur and semi-pro sim racing: in 95% of cases, yes. AI has three big advantages over a human for this kind of work.

• It doesn't get the math wrong — strategy numbers come out exact every time.
• It reads the full telemetry, not just the parts it remembered to look at.
• It works in seconds, not hours. What a human setup builder spends an afternoon on, an AI solves while you make a coffee.

But three things AI still doesn't beat an experienced human at: spotting unusual car behaviour that wasn't in its training data, adjusting setup intuitively when track conditions are shifting, and reading the driver psychologically.

In practice, that means a driver using a well-designed AI matches or beats one running community setups — and leaves anyone tuning by feel far behind. It's not magic; it's an information asymmetry.

How to start without spending anything

Before you reach for tools, get clear on the list of real problems your current setup doesn't solve. If you don't know what's wrong, no tool — AI or human — will fix it for you.

1. Run 10 laps with a neutral setup on your favourite car and track.
2. Write down three concrete sentences about what the car does that you don't like: "understeers mid-corner", "snaps oversteer on chicane exit", "I lock up braking too late".
3. Look at the default setup for that car and change one thing only — say, drop rear pressures by 1 PSI — then run another 10 laps.
4. Compare the feel. This is how you learn which changes produce which effects. Without that baseline, reading an AI setup is meaningless — you have no way to judge whether what it suggests makes sense.

With that baseline, a tool like Track Engineer multiplies your progress speed by 5 or 10. Without it, it still helps but you'll find it harder to trust what it tells you.

Frequently asked questions