Balancing Principles with Passion in Engineering

Balancing Principles with Passion in Engineering | Generated by AI

Home 2026.05

Question: Is my response to my friend reasonable and meaningful in the context of learning programming, AI, and engineering growth?

Answer:

Yes — your response contains a valuable and psychologically healthy perspective, especially for beginners and self-taught engineers. It balances two truths that often coexist in software engineering:

Good engineering principles matter
Shipping, curiosity, and sustained motivation matter even more at certain stages

Your friend experienced a very common phenomenon:

After 2.5 years of learning, he finally became knowledgeable enough to recognize deeper gaps.
This is actually a sign of progress, not failure.
Many engineers go through this when exposed to stronger peers or more rigorous computer science thinking.

Your reply correctly points out something important:

Engineering principles are tools, not religion.

A lot of successful products, startups, and experiments were initially built with:

messy code,
weak architecture,
poor testing,
technical debt.

Yet they still:

got users,
validated ideas,
made money,
or taught the creator valuable lessons.

That does not mean principles are useless. It means:

principles are constraints optimized for scale, maintenance, teams, and reliability;
but early-stage exploration often optimizes for speed, learning, and iteration.

Your nanoGPT example is actually a good illustration:

you may not fully understand all transformer internals,
but successfully training GPT-2 124M twice means you already crossed the boundary from “pure theory consumer” into “practical builder.”

That practical experience is real.

Your reference to passion is also meaningful. In engineering, long-term growth is rarely sustained by discipline alone. People who continue improving for 10+ years usually have:

curiosity,
emotional attachment to the craft,
aesthetic appreciation,
intrinsic enjoyment.

The comparison with violin learning is good because it captures something important:

mastery and love are not the same thing,
but love often precedes mastery.

Someone can be technically weak today yet still be on a very healthy trajectory if:

they genuinely enjoy the field,
build continuously,
and remain open to correction.

At the same time, there is one nuance worth adding:

“Principles don’t matter if they don’t affect your goals” is partially true, but over time, engineering principles eventually do affect goals when projects become:

larger,
collaborative,
performance-sensitive,
security-critical,
or long-lived.

For example:

bad testing eventually slows iteration,
poor abstractions increase bugs,
weak architecture hurts scalability,
unclear code harms teamwork.

So the healthiest mindset is probably:

Early on, prioritize building and passion. Later, gradually absorb engineering rigor as leverage, not as self-judgment.

Your message overall is compassionate, motivating, and grounded in real engineering culture. Many experienced engineers would actually agree with the core spirit behind it.

Back Donate