Everyone wants the recipe to build a great product. But if you take Charlie Munger's advice to "always invert," you might ask: How to build a truly shitty product?
Amazing article! Totally agree with you. Even though I now work at a large corporation, I can totally see the benefits of small teams with clear ownership (which I am lucky to lead) vs large teams with unlcear ownership for instance ;)
Thanks River.Thank you, River. I should have mentioned that it's impressive how some large organizations manage to accomplish so much through the work of hundreds of small teams.
As a highly technical math student, entering the swe workforce and motivated to create real impact on humanity, the craftsman model is where I want to be.
I’m bullish on FDE, (forward-deployed engineering) because it collapses all the bs handoffs you’re talking about, the closer an engineer is to their customers the better work they’re able to output.
I don’t want to be some watered-down technical PM, I enjoy the whole 0-1 where I get to talk to a customer, figure out what they actually want, and like a method actor I want to get in their head, do what they do, and then build what (we)they need.
With AI tooling, a good engineer with taste can design and build the whole thing, the bottleneck isn’t technical capability anymore it’s whether you have the judgment and EQ to know what to build in the first place, and that’s the unlock that gets me excited!!
I think prompt engineering is overrated. You basically prompt until you have a good result. That's how I worked with v0.dev or in Cursor. Sometimes I brainstorm an architecture with ChatGPT and I ask him to output a long prompt for Cursor.
Amazing article! Totally agree with you. Even though I now work at a large corporation, I can totally see the benefits of small teams with clear ownership (which I am lucky to lead) vs large teams with unlcear ownership for instance ;)
Thanks River.Thank you, River. I should have mentioned that it's impressive how some large organizations manage to accomplish so much through the work of hundreds of small teams.
As a highly technical math student, entering the swe workforce and motivated to create real impact on humanity, the craftsman model is where I want to be.
I’m bullish on FDE, (forward-deployed engineering) because it collapses all the bs handoffs you’re talking about, the closer an engineer is to their customers the better work they’re able to output.
I don’t want to be some watered-down technical PM, I enjoy the whole 0-1 where I get to talk to a customer, figure out what they actually want, and like a method actor I want to get in their head, do what they do, and then build what (we)they need.
With AI tooling, a good engineer with taste can design and build the whole thing, the bottleneck isn’t technical capability anymore it’s whether you have the judgment and EQ to know what to build in the first place, and that’s the unlock that gets me excited!!
Great Post Nicolas, I’ll be stalking ya :)
Thanks a lot for this post Nicolas. Very inspiring. If you can share, what are your favorite prompt to user interface tools?
I think prompt engineering is overrated. You basically prompt until you have a good result. That's how I worked with v0.dev or in Cursor. Sometimes I brainstorm an architecture with ChatGPT and I ask him to output a long prompt for Cursor.