Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Should I learn to code (again)?
7 points by dgrayman 6 days ago | hide | past | favorite | 6 comments
Hi HN / long time reader, first time poster. I've tried to learn to code by myself many times. Got someplace intermediate each time, but couldn't be consistent enough to take it to the next time and make it a life skill.

During that process, learned a lot about underlying technology and how to evaluate them, use them, and work with engineers.

Now my day job is a VC, side hustle is basically a PM on a machine learning project.

Am frustrated that I've come as close as possible to working in tech without actually doing it. And feel powerless because building cool things yourself seems much more fun than talking about them or analyzing them.

The best convos I have during the week are always founders and hackers, not VCs. And the best VCs are ex-founders or ex-operators.

So 2 questions:

1. Should I try learning to code again?

2. How should I go about this to maximize chances?






1. What's your goal of learning to code? If you want to be a professional software engineer (which you may or may not need more data to decide), then perhaps some kind of education is necessary. Bootcamp can work. A completely new undergraduate obviously works. If your education is close enough maybe you can get into a CS/SE masters program and go from there.

2. If your goal is to have deeper technical conversations with engineers, perhaps you can ask to "pair program" with them if there are engineers at where you work and you'll learn about what their jobs involve.

3. If you just want to be more of the engineers community, then I'd say being able to actually code is not required. You can go to a bunch of different events to get to know more of them. Hackathons, various career/hiring events, board game sessions etc etc.

4. If your goal is something else, then there are other options. There's codacedemy, there's udemy, there's freecodecamp etc. If you want to be more hands on, maybe you can join one of your engineer friends on one of their side projects and let them teach you in that concrete situation.


Thanks for the reply.

Have a lot of engineer & founder friends I talk to regularly. I think overall goal would be to go one step up from connecting and actually be a peer in some respects.

Build up skillset to be a founder-operator in the future. Along the way, I'm sure it's helpful to the day job in investing.

Re: 1. Definitely not this. I've realized my skillset / interests don't coincide with writing code at Big_Tech for a living.

RE 4. Have tried a lot of those programs. They don't click after the beginning stages, and don't do a good job graduating you or tracking you. Also intrinsic motivation is a problem -- i'd like to take a half year off to just do that.

RE 2 and 3: These are good ideas. Have been doing them to some extent. But taking next step of being in community -> contributing to it.


One approach that would deepen your technical proficiency and also have side benefits for your day job would be to dive deep into a particular area of interest, research it's origins and then analyze/criticize/evaluate the current state of the art.

Yes. It is clear you must do this.

The question is how to start.

1/ Build something that solves a problem. Solve your own problem. See if there is some manual process you can automate.

2/ Get a job where you will be exposed to a lot of interesting problems. Like technical consulting.

3/ In your VC gig start expressing your ideas through programs. Embellish your slides with working prototypes.

4/ When you’re building things try going end-to-end as quickly as possible. Don’t embellish. Don’t optimize. Just get to the finish line as quickly as possible. This process makes you hit problems and roadblocks quickly so you have more time to think and to come up with good solutions.

5/ Programming is about solving problems. Try to maximize your problems and your learning will maximize.


It is a double edged sword. It can look cool building tools but there are downsides as well. If you want to learn to code then start by solving a problem you have. Even better get someone to look over it for you so that you get feedback in order to get better. There are a lot of great reddit groups that can help with this.

1. why not? 2. start building



Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: