HN user

AwesomeFaic

95 karma
Posts4
Comments90
View on HN

Take what you want to learn and break it apart into core components. Break those up. Break those up. Get to a granularity where it doesn't make sense to reduce concepts further. Prioritize these bite-sized items. You don't have a single thing you want to learn, you have a list of things you want to learn. This refinement process shouldn't take more than a few days, and each individual item shouldn't take more than a few days. As you begin crossing things off your list, the "few days" will turn into a week. Then a couple weeks. Then a month. Soon, you'll be seamlessly transitioning from one concept to another without realizing, as it will come naturally since the list was already ordered in an intuitive manner (and you're far more motivated and seasoned at this point). When you're done with that list, start a new one. Follow the same process, but it will ramp up far more quickly. Repeat until you no longer need the process (but if you enjoy the process there's nothing wrong with it)!

I try to find micro-moments throughout the day at work, either sitting at my desk, alone in an elevator, in the bathroom, etc. Nothing serious, but a bit to regain focus & center during turbulent days.

I always start with a casual phone call to gauge their style and personality, and to gauge how well they understand the challenge. If it goes well and there's a spark, I reach out with a custom statement of work based on notes I took during the initial call, and we formalize terms, budgets, etc. to fill in the blanks. That almost always ends with a digital handshake by the end of the call and I start on the work immediately.

In a meeting and a half I understand the problem, their understanding of the problem, their personality and work style, and specific terms.

One I'm very interested in that I don't have the time or expertise to really tackle is single-lens photogrammetry (or photogrammetry in general). Right now there isn't a great consumer-friendly option that doesn't require custom hardware. Many online resources recommend specific lighting and high-contrast colors to get a barely-accurate scan.

Another similar-ish one that other people have started addressing is building a system that reliably detects photo/video manipulation (ie. photoshop & deepfakes). An interesting extension of that would be detecting audio manipulation in a video.

For my day job, I'm not sure. I do software development for insurance. It's a very multi-faceted industry, more so than the average person would expect. From a tech side the biggest issue is updating old systems without breaking integration with a million things across the board.

In my side work, I run an agency that aims to reduce the amount of BS in advertising (I know, a lost cause). We focus on being honest, creative, and making people happy, instead of falling in line with the rest of the "you're inadequate but this product will fix you" approaches.

Seconding this, Spotify at work is blocked and lots of active YouTube activity would look bad, so I fire up the beloved YouTube 24hr lo-fi beat stream and let it run all day pretty much

When I started my first consulting agency it was primarily focused on helping game devs treat their projects like a real business, which I witnessed was a common point of failure for indie studios. I reached out to my old Preproduction in Game Development professor (Brian Sullivan of Age of Empires fame), who also had a lot of consulting experience, and thanked him for everything while asking for advice moving forward. He seemed genuinely appreciative of the message and gave me some great advice.

I'm trying to do a bit of that, but it's understandably difficult. I have an easy day job time-wise, but even then I feel like I rarely do my best passion project work at home. Lately I'll wake up at 5 and work, go to the office from ~8 to ~4, and then get a couple more hours done at home afterwards.

I was at a company for 6 months that started with 9 hour days and after 4 months turned to regular 10-14 hour days. The worst was a week of 10+ hour days straight, usually there'd be one "normal" day each week but something was always going wrong (having to meet unrealistic deadlines, systems breaking, etc). Glad that experience is in the rear-view, it taught far more on self respect/care and connected me to a lot of great people (who also left soon after I did).

Edit: Forgot the initial question. It was a huge negative influence on an existing long-term relationship. If I stayed at that company it likely would have ended the relationship.

As a developer, practicing articulation and interpersonal skills has advanced my career more than any specific technology. I've gotten jobs because I was the most memorable and pleasant to be around, more so than explicitly being the strongest developer (although I obviously keep relevant skills up to date, too)

Depends how much money you can invest right now, how much time you have to spend, and how quickly you need to start hitting that $2000 mark.

Need it now? Check out Craigslist gigs (help with moving, quick jobs, website help), or try flipping thrift store and eBay stuff. Pros: Quick, no skills needed. Cons: Unreliable and more time consuming.

Need it reliably? Work on a business and work your ass off in your spare time. Pros: Easy to make more than $2000/mo if it goes well. Cons: Will take a lot of time and effort to maybe work out for you.

Alternatively, invest in a property. Pros: Easy $2000/mo through rent. Cons: Lots of upfront investment & research needed to not screw yourself over as a property manager.

Personally I'm in a similar situation and currently selling on eBay and running a side business. Been at it ~6 months and make 200-400/mo on top of my salary, but looking to ramp up the business more this year.

Edit: One option I overlooked of course is investing. The safer the investment, the more up-front capital you'll need and nothing is truly 100% safe. Gamble on options but be prepared to lose everything, if you really need something immediately.

I got an Ender 3 3D printer for last Christmas and while making gadgets and toys and gifts has been fun, it's fundamentally reawakened my interests in being creative again. I've learned new skills with silicone and resin, gotten back into 3D modeling (and finally learned Blender), discovered a bunch of new visual artists, and found new online communities to inspire me after a long day looking at code.

I'm new to ML and seeing approaches & explanations like yours is really helpful. Currently in an Introduction to Deep Learning course on Coursera and reading through notebooks outside of a strictly academic setting is enlightening. Thanks for sharing

My long-term hobby is building a home museum with a tri-fold emphasis on natural history, the history of mankind, and pop culture. I try to avoid the obvious artifacts and look for specific and obscure objects that highlight thought-provoking moments in life & history.

A few examples: Fulgurite, Manhattan project glass, Japanese Pesos, Nazi-Occupied Polish stamps, 1996 Olympic medal, bone records, DEA task force patches.

Also plenty of fossils, meteorites, international currency, memorabilia, etc. Inspired by a childhood spent exploring museums and, in recent years, Mini Museum on Kickstarter.

Edit: (Also many other less cool hobbies with tech & art)