Ask HN: What to do as a software engineer that cares more about what to build?

#1

Post by oncethere » Tue, Jun 08, 2021, 7:12 PM UTC

I've been an engineer for many years and it's become pretty clear that What a team chooses to build (or not) has far more impact than the architecture / code design choices. Nowadays the role feels like someone else figures out what to build, and we just code it up.

I love coding, but I've gotten to the point where I just can't enjoy it if I'm not involved also choosing what we build. The most interesting time in my career was when I did the spec, architecture, and code of a pretty fundamental system (small team, just lost the PM, fell to me as tech lead). The worst time was contributing for a number of years to a large software platform that had incredible groundbreaking tech and fundamentally failed in the marketplace.

It seems my only choices are to 1.) be a PM and stop coding or 2.) find a team where engineering also makes product decisions

Has anyone else successfully navigated this dilemma?

PS I've had the same experience as an engineering manager - less coding, no real responsibility for what the team chooses to build.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#2

Post by oncethere » Tue, Jun 08, 2021, 7:18 PM UTC

Does anyone else feel this way? Curious about your experience and what works for you. I know lots of engineers that are super happy working against solid requirements.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#3

Post by tppiotrowski » Tue, Jun 08, 2021, 7:22 PM UTC

Get in the habit of asking: "What is the problem?"

CEO - "We need a live streaming video platform"

Me - "What is the problem?"

CEO - "I want to do a webcast on our website"

Me - "Would embedding a YouTube video player work? It does x, y and z and it can be ready by tomorrow"

CEO - "Oh! I hadn't considered that"

You need to get back to first principles. Usually your boss wants to accomplish something and then they come up with a solution. Then they fall in love with the solution, not the problem.

You should be asking your boss what they're trying to accomplish instead of letting them propose the solution. If you help devise the solution/product you'll be much happier building it.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#4

Post by oncethere » Tue, Jun 08, 2021, 7:25 PM UTC

tppiotrowski wrote: post #3

Get in the habit of asking: "What is the problem?" CEO - "We need a live streaming video platform" Me - "What is the problem?" CEO - "I want to do a webcast on our website" Me - "Would embedding a YouTube video player work? It does x, y and z and it can be ready by tomorrow" CEO - "Oh! I hadn't considered that" You need to get back to first principles. Usually your boss wants to accomplish something and then they com…

Haha. Yes, I do that potentially a bit too much. It always feels like pulling teeth -- like the official process doesn't have room for that kind of back and forth. For example, in scrum, the product owner is meant to have that conversation, not the engineers.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#5

Post by markus_zhang » Tue, Jun 08, 2021, 7:26 PM UTC

I think Paul Graham has an excellent article named Hacker and Painter regarding this. I think the only way is to develop your own products.

So far I habe worked in a few companies as a business analyst and a bi developer but I have never seen one that allows the dev team to have some final say in the products. Products are always dictated by business team, then project manager or business analyst breaks them down to tickets that you guus understand, i. e. the scope. The lead further breaks them dowm into sprints and tickets.

So you can see that sometimes even product managers don't have much a say in the product.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#6

Post by oncethere » Tue, Jun 08, 2021, 7:34 PM UTC

markus_zhang wrote: post #5

I think Paul Graham has an excellent article named Hacker and Painter regarding this. I think the only way is to develop your own products. So far I habe worked in a few companies as a business analyst and a bi developer but I have never seen one that allows the dev team to have some final say in the products. Products are always dictated by business team, then project manager or business analyst breaks them down to…

Thanks for the pointer to the article!

Yeah, that's a good point. The PM is sometimes pretty constrained by the business. Maybe it's just a grass is greener thing. Have you been anywhere where the eng team had some say in what they make?

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#7

Post by markus_zhang » Tue, Jun 08, 2021, 7:41 PM UTC

oncethere wrote: post #6

markus_zhang wrote: post #5

I think Paul Graham has an excellent article named Hacker and Painter regarding this. I think the only way is to develop your own products. So far I habe worked in a few companies as a business analyst and a bi developer but I have never seen one that allows the dev team to have some final say in the products. Products are always dictated by business team, then project manager or business analyst breaks them down to…

Thanks for the pointer to the article! Yeah, that's a good point. The PM is sometimes pretty constrained by the business. Maybe it's just a grass is greener thing. Have you been anywhere where the eng team had some say in what they make?

Sorry I have never seen one such place. I mean for sure in each place I stayed (custom service center, gaming and one I can't say), the eng team has SOME say in the product, but most od the time it's just small details of the design or something similar that they can argue about. Actually data analysts have a lot more say because they have the data to back them up when they want to make recommendations.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#8

Post by quickthrower2 » Tue, Jun 08, 2021, 8:18 PM UTC

tppiotrowski wrote: post #3

Get in the habit of asking: "What is the problem?" CEO - "We need a live streaming video platform" Me - "What is the problem?" CEO - "I want to do a webcast on our website" Me - "Would embedding a YouTube video player work? It does x, y and z and it can be ready by tomorrow" CEO - "Oh! I hadn't considered that" You need to get back to first principles. Usually your boss wants to accomplish something and then they com…

Never that simple at a macro level. Business will claim more knowledge of the “market” and are probably right if they have got feedback from customers. “We need to build X as part of our strategy”

Sure you can choose to use nodejs instead of ruby or embed video instead of streaming it yourself or find a better confit system than admin pages but the entire product ideas are probably decided by the business well in advance of when the developer gets whiff of it.

Imagine you are about to purchase a car and the car dealer weirdly tries to convince you to take up cycling via “what’s the problem”. You’d probably walk to the next dealer as you’ve made up your mind you want a car.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#9

Post by quickthrower2 » Tue, Jun 08, 2021, 8:21 PM UTC

oncethere wrote: post #2

Does anyone else feel this way? Curious about your experience and what works for you. I know lots of engineers that are super happy working against solid requirements.

I do too. There is something about building your own idea that’s magic. Even partially magic is given the “we need an x” but deciding on the technical and visual design. Team work is probably more effective here: graphic designer. UX, product managers etc. at producing the polished results we expect in the 2020s but man … why are we drawn to side projects? The urge to build from the ground up is strong.

Re: Ask HN: What to do as a software engineer that cares more about what to build?

#10

Post by tppiotrowski » Tue, Jun 08, 2021, 9:24 PM UTC

quickthrower2 wrote: post #8

tppiotrowski wrote: post #3

Get in the habit of asking: "What is the problem?" CEO - "We need a live streaming video platform" Me - "What is the problem?" CEO - "I want to do a webcast on our website" Me - "Would embedding a YouTube video player work? It does x, y and z and it can be ready by tomorrow" CEO - "Oh! I hadn't considered that" You need to get back to first principles. Usually your boss wants to accomplish something and then they com…

Never that simple at a macro level. Business will claim more knowledge of the “market” and are probably right if they have got feedback from customers. “We need to build X as part of our strategy” Sure you can choose to use nodejs instead of ruby or embed video instead of streaming it yourself or find a better confit system than admin pages but the entire product ideas are probably decided by the business well in adv…

Maybe it's more of a startup thing. Another startup I worked at turned their entire website into an SPA over the course of 12 months. The website was never broken. It wasn't sleek but it worked. I kept my head down and rewrote it.

The email order confirmations, however, took 30 minutes to be delivered to customers. In the end, it was this uncertainty of whether an order went through that killed the customers trust, not the fact that it was not an SPA. I wish I had asked more questions instead of blindly following orders.