Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: What skills would you invest in learning?
67 points by chakkop on July 27, 2015 | hide | past | web | favorite | 66 comments
I'm considering taking time off work (up to 6 months) to invest in learning some new skills. I'm 28, have a solid background in math, and my professional experience is mostly in finance (sigh). What are some useful (technical or non-technical) skills that it's worth investing into learning?

When I look around me at conspicuously happy people I see a) cooking, b) exercise (less of a skill than a lifestyle but there you go), and c) a wide, wide variety of X plus Y such that you're better at Y than anyone who can do better at X and vice versa, for X and Y which are chosen to be commercially valuable.

Professionally, if you want a skill with a stupid amount of leverage, public speaking is pretty high up there for white collar workers.

If you're technical there are many, many options for adding one more arrow to the quiver. My specific recommendations would depend on your career goals, but for generic HNers, mastering the deployment story for your stack of choice (DevOps is more than a bit buzzwordy but some of the tools available now are just fantastic -- Ansible, Consul, Docker, the AWS stack, etc) would be up there. For people who are more comfortable on backends, learn React, which will take you through a nice swathe of modern front-end tools and practices. For folks who work more on the frontend, maybe Rails if you see CRUD/apps in your future or Go if you enjoy systems programming more.

The X and Y analogy reminds me of a tweet from Josh Willis about Data Scientists [1]

"Data Scientist (n.): Person who is better at statistics than any software engineer and better at software engineering than any statistician."

[1] https://twitter.com/josh_wills/status/198093512149958656

Could you please given an example or two for X + Y? I am not following that.

For me the original plan was bilingual (English/Japanese) plus engineering but I ended up getting more career mileage out of being able to program rings around most SaaS companies' marketing teams. (A high bar that is not.)

You can pick many, many things here, though. Combining programming (any stack which lets you ship things) plus any other white collar profession works well, too. Try embedding in anybody's workday and just sit on your hands and watch the insanity as they do any data-processing work, for example. It's insane how much of day-to-day accounting work exists because of the lack of 50 lines of Ruby.

In my experience I've found that you can automate a lot of white collar work if the rest of the business will become organized enough to make that happen. If they aren't collecting or ingesting the right data already, then the hard part is changing their behavior, not writing the code.

could you elaborate on what you do when "programming rings around most SaaS companies' marketing teams" ? Thanks!

His blog does a really good job of explaining this. And you'll probably learn a lot of other stuff. It's worth getting sucked into kalzumeus :-)

Let's say, a coder with significant law knowledge, such that he knows more about law than all the other coders and more about code than all the lawyers, in a given company.

I think this is a good example. I would just add that it helps to be able to deflect when the lawyers come to you with all of their technology problems and coders come to you with legal problems (or whatever appropriate mix of skills and problems). You don't want to end up having to fix all the printers.


Sales. Learn to sell something. If you're taking time off, why not consider doing phone cold calling for a charity. This is particularly brutal but probably a sharp, free course in improving your sales and cold calling skills.

Everyone needs sales skills, whether it is to understand why people buy, or just to 'sell' ideas, whether pitching to investors or convincing a partner/superior or team to implement your idea.

On this note, design and web skills. Whatever you do, its useful to be able to make it look good. If you are working with people who aren't savvy with web tech then it is very useful to be able to knock good looking sites up quickly and will impress easily. learn how to use bootstrap templates to generate simple nice looking web pages. also bonus points for learning basic graphic design skills so you can generate your own images to customise your pages further.

But only after doing a few day course in how to cold call, so that you can write/tweak your own scripts, and/or know why the ones they give you are the way they are.

Also, doing various sorts of hard selling (phone, on the street, for charity or selling products) would probably teach more than doing charity sales for 2 months. I'm too chicken to do something like this myself though...

I appreciate the skills the Center For Applied rationality instilled in me most every day.

>Ever made a mistake? Missed an opportunity? Of course; but what’s interesting is how cognitive scientists have found even highly educated and successful people to make predictable errors in judgement, and just knowing about these experiments often isn’t enough to prevent these mistakes. It actually takes practice to form new mental habits. At our workshops, you can learn about newly discovered failure patterns in human decision-making, and begin training to overcome them…

Think of it as martial arts for rationality. You know about things like the sunk cost fallacy, but they do workshops to make sure you recognize it when it's happening to you.

A great bunch of very skilled people. The personal advice helped me get over some hangups about procrastination.

I think it's a great foundation to help learn other skills, and asses what information will be valuable to you.

Sounds interesting. Do they have a pdf or book out?

They have a collection of resources including recommended readings: http://rationality.org/videos/

If I had 6 months and no obligations, I'd learn to draw and to storytell, and I'd make and publish (for free if I can afford it) a graphic novel adaptation of Asimov's Foundation.

Current item 1 on my "to do once we exit" list.

It's a brilliant set of books by one of the most brilliant authors ever, and IMHO has a philosophy and structure that is rarely found in contemporary books or graphic novels. It would be a challenge to bring that vision to the paper well.

In 6 months, you could learn the basics of drawing. Enough to make a simple graphic novel, to have fun and get the experience. But something tells me the source material demands a more complex approach. The best illustrators take years to refine their technique.

Not trying to dissuade you. If you love drawing, absolutely do as much of it as you can. Just don't stop after 6 months.

Start anyway. The rough draft will always be a rough draft.

Asset allocation - so to speak - over one's personal time is usually a very personal thing. Six month is a long time, but it is not enough to get reasonable good in piano playing. Or to acquire a black belt.

If you have a bit of talent and help, you could get fluent in a foreign language in 24 weeks.

But then, there are some talents, that you wanted to foster during the last years, but just did not have the time. Maybe it's easier to start from there.

Anyway, some suggestions into the blue:

- read books and travel

- learn excellent writing skills by writing a short book about what you learned so far in math/finance

- move to a foreign country where you do not speak the language and try to survive (two of my friends did this, and they did well)

Code. Teach yourself how to code. Create a project that you would find useful. Given your finance background, create a system to manage your personal finances or automate some of the tasks you perform at work. You have the mathematics background, so logic isn't a hurdle. If the glove fits... code.

I second this. You are after all asking the Hackernews community! Like what dankruss says, you have a lot of complementary skills too.

I'd recommend some sort of dynamic web based language so you can launch your own startups in the future if you want.

Which language though? Coming from someone who knows a bit but doesn't work in the field.

I would say Python. Easy enough to be a good teaching language + a huge ecosystem for real work, spanning everything from web dev to machine learning.

I second the other suggestion - Python. I think its a great language to learn first up. I feel its biggest strength is its excellent standard library and ecosystem Python allows you to do really cool things with minimal frustration. This is important because once a person gets past the basics what keeps them hooked is doing cool things what they've learned.

Its gotta be JavaScript. Given Meteor, JS is all you need to get an idea out the door.

To get an idea for a web site out the door. Not all ideas are web sites.

I'd recommend putting some time into learning how to change your body composition, improve your fitness, and control your mood. Better physiology improves all other aspects of one's life. Also practical psychology (why people do what they do, etc) is helpful. After that, learn how to create software.

Which subjects/areas of psychology would you recommend for someone looking for learning about practical psychology?

Here are several books I've found extremely useful. Ranked by how important I view them.

1. Thinking Fast and Slow - Daniel Kahneman. If you're going to read one, read this. A lot of theory undergirding how people think. Decision making by people will make a lot more sense after this.

2. Influence - Robert Cialdani. Less theory and more pragmatic advice on how to influence people.

3. Drive - Daniel Pink, Switch - Chip Heath, Made to Stick - Chip Heath, Predictably Irrational - Dan Ariely. Books on specific sub-categories. More pop-psych. Information density is less, but easier to read.

4. Poor Charlie's Almanac - Charlie Munger. Best known as Warren Buffett's partner, this book is a collection of his speeches, letters, etc... You get an idea of how he thinks but, but you have to dig through the repetitive ramble to get it. Think of it as Charlie observing a lot of the prior principles but putting it into a real life/business context.

Thanks for the list! I had #1 and #3 in my list already, I'll check the others too :).

Why "finance (sigh)"?

I'm aware HN doesn't have the highest regard for the finance industry, but I'd like to hear your perspective: What makes you not so proud of your current skills?

I think "sigh" implies low work satisfaction. OP didn't come here for a therapy session, he wants to learn new valuable skills!

OP where are you at? I've consulted in a variety of different places in the financial industry, the opportunities range from terrible, boring and soul sucking, to extremely profitable and very interesting - the financial space is actually quite diverse

I think the skills are OK. I find the average personality type I encounter in finance a bit abrasive: one-dimensional, sharp-elbowed, too full of certainties. I worry that this might be a general phenomenon in the corporate world :/ (of course, these people also tend to be much much happier).

The 'sigh' was almost reflexive: even writing the word 'finance' makes me feel bored.

> "too full of certainties"

This is one of the worst possible qualities to have in finance... well, unless you're on the sell-side.

But all too common on the buy-side, e.g. in PE/VC, and even most public equity investors.

Try building a computer from first principles[0]. You'll learn logic, circuits, systems programming, etc. Everything that runs our lives.

If you have a solid background in maths perhaps look into picking up a semester in some theoretical maths. You might develop a few hunches that could turn into innovative ideas when you return to your job.

Work on an open problem in maths with a small prize. See if you can crack it[1]. You could end up with a proof, some pocket change, and your name in the history books. At the least you'll learn what it takes to get there.

Pick up a few online courses or books on machine learning and dig through the Kepler data archive[2]. You'll learn statistics and modelling and perhaps confirm valuable research results! Space!

Go plant some trees. It's a small favor for the environment. It's hard work and you'll get really dirty. This is its own reward.

Volunteer in a foreign country. You'll meet new people and introduce yourself to problems you would never have considered in your lifetime.

[0] http://www.projectoberon.com/ [1] http://www.numberphile.com/videos/happy_ending.html [2] http://kepler.nasa.gov/Science/ForScientists/dataarchive/

Depends on what your goal is. If you want to make a lot of money, I'd say learn about an industry, especially one that's in a bad need of a technological overhaul. Make connections with people who can help you shape and advance a product. Almost nobody ever made money creating TODO apps, but plenty of people made loads of money bringing technology to where it was absent.

As for specific skills, sales and marketing are great ideas. If you want to be a very desirable coder, I'd say learning two technologies that don't normally go together but are both in demand at all times. For example, I have met very few people who have equal passion and skill for developing for iOS and Android. Add Windows Phone for completeness and most small companies would hire you on the spot because they can't afford a separate developer for each platform.

Another example of that might be front-end and backend work. Basically, if you can code up your own device drivers, then quickly create beautiful browser or mobile interfaces, you have become a truly full stack developer.

The ultimate triple threat would be: backend, front-end (web + mobile), and design. People that can quickly design beautiful things, then implement them are incredibly rare.

While we are all here, lets make a list of some industries that are very badly in need of a tech overhaul. I'll start:

Any sport where girls are required to put bows in their hair. Ballet sure could use some carbon fiber thingy in those shoes. Same with figure skating. Gymnastics is a bit better, but I'm sure there is room for something.

Pipe and cigar smoking. The vaping trend hasn't hit them yet, but it will.

Road construction. It is terrible! Potholes everywhere!

I agree with others here who recommend learning how to code.

But instead of chasing the latest trends in frameworks, I would recommend learning ruby on rails instead. It's a mature framework that is still very popular. Learning it instills good principles that you carry over when learning future frameworks.

The best way to learn rails is: https://www.railstutorial.org/book

It's written for absolute beginners. Work through the entire book, and you'll have a good foundation for web development.

That's a question you have to answer yourself. It sounds like you're dissatisfied with your finance career. If that's true, don't make the same mistake again. Don't try to figure out what skills are worth investing in (in other words, don't try to calculate ROI). Work on whatever you're passionate about (and if you aren't sure what that is, work on whatever peeks your curiosity).

It's an interesting question, and maybe a wrong question altogether. From my current experience I've noticed that it does not really matter what you learn as long as you learn something. To go event further, choose something unrelated to your current skill set. When you learn something new, the way you perceive the world, the way you think in general changes because you have more variables to work with. You will start observing things that were always there, but the subconscious part of your brain was filtering it out because it was considered "useless". You will understand more, and at some point will be able to make connections between apparently unconnected topics, get new ideas, innovate. But if you want to learn a new skill just to monetize it (get a good payed job quickly), things are a bit different. Then you must stay in your area of expertise, and choose something close to that, something that either has potential in the near-future, of it can be monetized now. You will have to make a bit of research on that one depending on the area you live in and the number of open jobs available.

I just did this actually. I ran a startup for 2.5 years as the CEO, and when we ran out of money, I moved to London for 3-4 months and joined the code school/coding cooperative Founders&Coders. Never learned as much in so short time ever.

It's probably best to try and learn something that you get excited about learning. You're going to be your own motivator, and if you give up half ways, it's probably not worth it.

Another tip is to try to get in touch with a community of people you can learn together with. After I moved away from London, I've learned a lot less, even though I have just as much time. I'm just not around as knowledge hungry people. This makes all the difference.

Another technique is to try and devote one week to a subject and check if you like it. I did it with machine learning: https://medium.com/@oslokommuneper/machine-learning-in-a-wee...

You can try coding, design, marketing etc. Whatever you like. Worst case, you waste a week. Best case, you discover a new passion.

Write a book.

While it appears to be an "activity" rather than a "skill", I believe it is a skill.

My recommendation is write on something which you know well; most probably in finance as you appear to be most experienced in this field. It will be even better if you can make it programming oriented, as there are a dearth of programming books specific to specialized domains.

1. Writing well is something which sets apart great technical people from the good technical people.

2. Writing is much more difficult than it sounds. It requires much more clarity of thought than when you are working. At work, some things may be obvious to you - only when you write about it do you think about the fundamentals of a domain.

3. Depending on your goals and the medium of publishing, it may teach you something else - like HTML or Latex, for example. It may also lead you to explore options for self-publishing a book, which can be a good skill later on.

Sales, marketing, and JavaScript. In that order. Then you'll be unstoppable in anything you decide to do.

O tempora o mores. :)

If I had six months I'd deep dive into ML and AI. They're about where the Internet was in the late 70s/early 80s. A lot of people have a sense they're going to be very important in the future, but most of that future hasn't been invented yet.

Third ones covered, what are the best resources for the first two?

Here are the resources I've found most helpful:

Sales -------------

Steli Efti: his YouTube videos + blog posts at blog.close.io

Smart Calling by Art Sobczak

Question Your Way to Sales Success by Dave Kahle

Marketing -------------

Traction by Gabriel Weinberg

To add,

* Good regexp chops in a scripting language.

* Cooking, because it's a proxy for skills in procedures, taste and aesthetics, and makes you more able to keep a healthy diet.

* How to set up, maintain and use a commuting bicycle. Easier and more productive than the gym.

6 months is a long time. If I were to do the same, here's what I would work towards:

- Guitar

- Programming language in a paradigm I don't currently work in

- Join a hackerspace and build something small but new every few weeks

- Wildlife Photography

I spend a lot of time in the computer, and while I would not want to give that up, given the time, I'd like to pick up some skills which have bugger all to do with using a computer in a way that I'm used to. I'm already doing two of those things in my spare time, but if I had 6 months, I'd make those my primary focus, and add in the others as my "hobbies".

Your question is hard to answer because it's not clear what your goal is. More math is always useful, but you wrote you already have a solid background there. Public speaking, communication, and "people skills" in general are always good to have and might become more important later on in your career.

If you want a fun list of topics, here are some off the top of my head that i might look at in your situation:

- fully homomorphic encryption

- recurrent neural networks & machine learning

- proving computer programs correct & computer-aided mathematical proofs

I'm surprised that nobody advised to learn algorithms. It's not only fun but also algorithmic skills offer bright career perspectives.

Top-tier employers requires deep knowledge of algorithms and math rather than concrete technologies (in case of average employers it's quite opposite).

If you strong in math, it would be easy for you.

In general, if I had solid background in math, I would consider to join research team.

I come from a very technical background--CS with Machine Learning, and I'm currently diving into the humanities and arts. I'm using Hackdesign as a primer into the arts and design frame of mind.

I think it's best to engage in domains of thinking that are worlds apart from where you currently are.


Not buying one and flying it.

Building one.

There are a lot of guides online. People online will help you. They are exciting. They teach you about a lot of skills from soldering to robotics to design to flying to rigging cameras.

They are exiting and you will end up with a nice product you can keep forever or give as a gift.

> learning some new skills.

I would say whatever sparks your curiosity. In most cases and somehow it would fit in your daily tasks.

For what purpose? To advance your career? Self-improvement? Fun?

1) I'm guessing more finance and math.

2) I second the CFAR recommendation.

3) Whatever you want!

I want to do about 90 percent of what the comments on this page say. Very good ideas...so little time...

how to help people --- you learn to do that and a lot of other stuff will fall in place.

Three most profitable things I've learned: writing, sales, and machine learning.

I'm curious - are there resources you'd recommend to bone up on sales?

Learning to write embarrassingly clearly.


nonviolent communication (cnvc.org)

machine learning and data analysis obviously!

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