Mr Z

AI Prompt to Get Back in Shape: By an 18-Year Coach

Francesco Zen, founder of ZTraining · Published 9 July 2026 · Updated 9 July 2026

The prompt

Copy-paste prompt, built from 18 years of coachingAct as my coach for setting a calorie deficit properly, not a crash-diet generator. Before giving me anything, ask me for: my height, current weight, a rough body-fat estimate, my activity level, how many days a week I train, and how quickly I am hoping to lose weight. Once you have that: 1. Estimate my maintenance calories from my stats and activity level, and show me how you got there. 2. Set a moderate deficit from that maintenance figure, not an aggressive one. Tell me the rate of weekly weight loss this deficit should produce, and be honest if what I asked for is unrealistic or unsafe. 3. Protect my protein target first, using lean mass as the basis, and take the deficit from fat and carbohydrate, not from protein. 4. Warn me that a deficit which is too aggressive costs muscle, energy, and training performance, and is harder to sustain, not just harder to feel. 5. Tell me this deficit is not fixed. As I lose weight, maintenance drops too, so the same calorie target becomes a smaller deficit over time unless it is recalculated. 6. If I tell you progress has stalled, do not just cut calories further as the first move. Ask about training, sleep, stress, and adherence before assuming the deficit itself is the problem. 7. Be direct if my expectations for speed do not match a sustainable approach. Ask me the setup questions now, before giving me a number.

Why this prompt works

The most common calorie-deficit mistake is not a slow one, it is an aggressive one that costs muscle and adherence long before it costs meaningful extra fat. A deficit that protects protein and stays moderate holds onto lean mass and training performance far better than a steep one, which is why this prompt sets protein first and takes the deficit from the remaining macros.

Maintenance calories also drop as body weight drops, so a deficit calculated once at the start quietly shrinks in real terms over a multi-month phase. Fixed calorie targets stall for exactly this reason, and it is often misread as a plateau in the body rather than what it actually is, a target that needs recalculating.

A real-world example

Take someone hoping to lose 1kg a week from a starting maintenance in the low 2000s. That rate typically implies a deficit steep enough to threaten muscle and training quality, and a coach-built prompt should say so directly rather than simply computing the number requested and handing it over.

The honest answer here is usually a smaller weekly loss target with protein and training protected, which is slower but is the version that is still working in month three.

Where a single prompt hits its ceiling

Maintenance calorie estimates are exactly that, estimates, generated from formulas that carry real error margins even with accurate inputs. The number this prompt gives you is a sensible starting point, not a precise measurement, and should be adjusted based on what actually happens to your weight over the following weeks.

It also cannot diagnose a genuine stall. A plateau can come from adherence drifting, sleep or stress changing, or water retention masking real progress, and distinguishing between those requires watching a pattern over time, not reading one week's number.

Common questions

How big should my calorie deficit be?

Moderate, not aggressive. A deficit that protects protein and training performance produces slower but far more sustainable fat loss than a steep one, which tends to cost muscle and adherence.

Why has my weight loss stalled on the same calories?

Because maintenance calories drop as bodyweight drops, so a fixed calorie target becomes a smaller deficit over time. It can also reflect adherence drift, sleep, or stress, not just the numbers on paper.

Is a bigger deficit always faster fat loss?

Not reliably. Beyond a certain point a steeper deficit costs muscle and training performance faster than it adds fat loss, and it is harder to sustain long enough to see the result through.

Francesco Zen is the founder of ZTraining and ZLIFE, with 18 years of coaching experience working with executives across the UK, US, and UAE. He built Mr Z, an AI coach trained on his methodology to bring the same approach to a wider audience.