Project Sculpt: making health the default

If you looked only at the result, this could easily be a story about how I lost weight. Looking back, though, what I want to keep is how I spent six months finding problems, changing rules, and gradually making a rough weight-loss plan fit my own life.

I did not design this system on the day I started. I built it over six months, fixing one bug after another.

This is not a weight-loss prescription to copy. Chicken, sweet potatoes, and push-ups were parts of my particular version. What matters to me is how that version developed.

01 · Eating less worked, but it was not the whole answer

In March, my understanding was very simple: skip breakfast, eat lunch and dinner, reduce staple foods and carbohydrates, cut snacks, eat less overall, and do not count every calorie.

It came down to one sentence: I want to lose weight, so I need to eat less.

And it worked. My weight fell quickly in the first few months. That made it tempting to keep pushing in the same direction. If eating less worked, would eating even less work better? Did faster weight loss mean I was doing better?

The scale told me whether I was getting lighter. It did not tell me whether my body was getting better.

In June, I noticed that my strength and physical performance were poor. Basic activities such as push-ups and skipping rope made something clear: I had become lighter without becoming stronger.

Morning exercise entered the plan. It grew from basic conditioning into more complete strength training and badminton. For the first time, the project expanded from losing weight to looking after my body.

Later, I set priorities for the measurements: training performance first, then waist measurements and progress photos, followed by average weekly weight and BMI. If my weight was falling while my strength kept declining, I would no longer automatically call that progress.

02 · My body started sending warning signals

Early on, the logic was seductive: eating less works, so keep eating less. By mid-to-late June, the restrictions began to feel excessive. For the first time, I asked myself: am I being harsh with myself?

I had gone too long without eating meat normally, and kept wondering what else I could reduce. In early July, I felt noticeably dizzy on standing up. At the time, I thought it might be related to excessive restriction and too little carbohydrate.

That was not a medical diagnosis. But it was enough of a warning for me to reconsider my own plan.

I began easing the restrictions: more carbohydrate, normal portions of meat and protein, and no automatic further cuts. Later, I added a rule in the other direction: if training performance kept declining, or I felt marked fatigue, dizziness, or weakness, I would consider recovery and food before cutting more.

This changed the question I was asking. I had been looking for something else to reduce. I began asking:

How can I lose fat gradually while keeping my body functioning normally?

Those questions look similar but lead to different decisions. The second gives my body’s feedback a say.

03 · The plan has to leave room for real life

Six months is long enough for friends to invite you out, weekends to take you away, work to disrupt the schedule, and days when you really want something to eat.

At first, I easily called these things failures. Later, that word began to feel strange. Are friends, shared meals, weekends, and unexpected work interruptions—or simply life?

A plan that works only when there are no shared meals, no friends, no surprises, and perfect energy every day belongs in a spreadsheet.

So I formally made room for flexible meals. Eventually, both lunch and dinner on weekends could be flexible. More importantly, I stopped compensating afterward: no starving myself, no sudden extra exercise out of guilt. The next day, I simply returned to the normal plan.

A good system makes it possible for one mistake to remain one mistake.

In early July, I wrote “It turns out the hardest part is not starting” . I thought the last stretch of weight loss was the hardest. Looking back, I would add this: the difficult part is keeping the plan workable as real life keeps intervening.

Sleep also became part of the picture. Morning exercise, work, English study, and entertainment all competed for the same 24 hours. Late nights affected both training and energy the next day.

Some supposed problems with follow-through began to look like problems further upstream. If I went to sleep at one in the morning, demanded a good workout the next day, and then blamed my willpower when I failed, perhaps the starting point needed changing.

04 · What stayed with me was a cycle, not a meal plan

By late July and August, I was still losing fat, but my priorities had changed. Strength, body shape, and everyday energy mattered more to me than the number on the scale.

I no longer wanted to trade declining strength and feeling worse for a few hundred grams less. In the final week, I explicitly ruled out extreme dieting or dehydration to produce a more impressive finishing number.

In March, I asked how to get lighter. In September, I asked how to have a healthier body over the long term.

If someone now asks whether they should also skip breakfast, eat chicken and sweet potatoes, and do push-ups, I would tell them not to copy my version.

What I want to keep is this cycle:

Set a goal → establish defaults → act → look at feedback → find a bug → change one variable → try again.

Three parts matter most to me. Defaults should be simple enough that I do not have to decide everything again each day. A single fluctuation should not immediately change the plan; a sustained deviation deserves attention. And after a failure, the first question should be how to return.

This is why I increasingly stopped treating self-discipline as the main explanation. A useful system makes the desired action easier to do by default, instead of waiting for an exceptionally strong-willed version of me to show up.

05 · Rethinking “rely on yourself”

I used to strongly believe: rely on yourself, not on anyone else.

Looking back, damn, that sentence did me quite a bit of harm.

It could quietly turn into a strange kind of pride: I should already know the answer. If I cannot solve it, I am not clever enough. If I cannot keep going, I am not trying hard enough.

This experiment did not work that way. I used AI extensively, bringing it my data, training, food, and physical state. I asked what might be happening, whether earlier rules still made sense, and whether to keep cutting or allow more recovery. Then I decided what to do next.

AI cannot train or eat for me, or bear the consequences of a bad decision. It also makes mistakes, so the final choice remains my responsibility.

But it gave me a strong sense of something:

I have to do the work. The knowledge does not all have to come from me.

I do not need to reinvent nutrition or repeat every mistake other people have already made. For me, AI lowered the cost of reaching existing knowledge and getting outside feedback.

My understanding of independence changed too. I can actively seek useful knowledge and tools while remaining responsible for the choice I make.

Admitting that I do not know, cannot do something, or need help does not make me weaker. Sometimes progress begins precisely with that admission.

06 · Taking better care of myself

Writing this, I realize that these six months also became a concrete attempt to understand how to care for myself.

Always satisfying the present moment is not enough. Never exercising when tired, always eating whatever I want, or endlessly scrolling instead of sleeping can feel good briefly without making life better over time.

The opposite extreme is no better: eating far too little, pushing through feeling unwell, or punishing myself for one large meal.

I now think of it this way:

Care for the person I am now, while accepting some responsibility for the person I will be in the future.

Train when it is time to train. Add food back when restriction has gone too far. Enjoy a meal with friends, then return to the routine without shaming myself the next day. I want to judge those choices by how they affect life over time.

If I had to leave one sentence for this project, it would be:

Build routines that help me take better care of myself.

The approach does not end on September 19. Weight loss will give way to maintenance, then gradually to strength and body composition. Other problems remain, and the most obvious is sleep.

I know early sleep matters and poor sleep affects the next day. I have even written rules for it. But knowing that I should sleep earlier and actually putting down my phone are different things.

That is likely my next project: observe, record, find the problem, make rules, try them, fail, fix a bug, and continue.

I am less interested in suddenly becoming an extraordinarily disciplined person. I would rather solve one actual problem in my life every so often.

This time, it was my body. Next time, probably sleep.

Making health the default is only a beginning.