I feel like most of my work has been wasted throughout my career.

In my 10 years, I can look back and I think maybe 3 years of my work still exists today. A couple companies are not defunct. A couple projects were literally cut from under me.

Is this normal? Do most software engineers end up having a ton of their work wasted?

Absolutely normal, imo. I was shocked my (MVP level) solution was still the main production software for a giant retailer. Ive also written and thrown away far more code than has been deployed

Yeah, that sounds about right - A lot depends on what level you’re at and how “agile” the company you work for is. It’s not wasted per se - Most of the time other engineers build on your work, usually as bugs are found or the feature evolves. But yeah, products (and companies) also will die off.

Code merely solves a given problem at a given time. It’s not meant to last forever. Rejoice in the knowledge and mastery you have gained instead.

If your code was used at some point, did what it was supposed to do when it was needed, and paid your rent, how can you think it was wasted?

I would say one partial cure for this is to contribute to open source or if you are using open source in your projects see if you can get your company to allow you to contribute any fixes you have implemented up stream. Although at times rather than asking, I have fixed issues in a cpl of hours after work and pushed these up. But the simple fact is everything, even the most "advanced" projects eventually become obsolete, so I wouldn't worry about it, it is just part of working in tech.

Yeah it got me money, but I kind of feel like I'm scamming VCs, which isn't a bad. It's just I'm not really contributing to anything

Yeah I'd feel like my work is more important if I worked on Postgres or an Apache project. Basically anything open source. The problem is getting paid.

I struggle with this feeling a lot as well. I think it's partly due to labor alienation, at least partially.

Scamming stupid VCs is just a bonus.

It's also a public service, the sooner each VC is dead broke, the sooner their stupid stops making life worse for everyone else.

I'm always relieved when my code get sunset =]

Yes

A lot of my work has been making programs to help prove custom communication protocols, before they are used.

So I am making:

1. A simulator for some device that sends data, like a sensor of some physical phenomenon.

• The data would either be decided by the user or from logs of a real device. It could be some GUI that lets the user vary values on the fly

1. A receiver of said data in the given protocol. This would then decode and display the data, which can then be tallied with original input data.

These were essentially only made to show that the protocol works and then used to test actual equipment which would be fabricated by some other company and as per the contract, the devices would need to be able to work with the simulators.

So once the actual devices are ready and deployed, little programs have done their job and are no longer required.
They do end up being shown to future junior devs though, as examples to learn from.