1,278 karma · joined December 7, 2010
The most successful version of it when broadband really took off was torrents.
Then we got git.
Now we got cryptocurrency/blockchain.
I think by the end of this hype curve/cycle, blockchain will be relegated to decentralized finance, which will likely become an interesting part of the broader financial system.
I wonder what will come next. I don't see decentralized identity verification taking off. Public/Private Key encryption for ID means relying on users to not lose their private key...or storing it for them and backing it up in some form of centralized and managed storage (i.e. cloud). So not gonna happen.
This has lead to much larger "towns" and suburbs where the sense of community is essentially gone. At best you might know a few neighbors on your street. So risk tolerance has gone down for a good reason. And when you see a strange kid out playing and you don't know who's kid they are, you're going to naturally worry about them and just assume the worst.
Another factor too astutely mentioned by someone below is the car dependent nature of most non urban environments, especially in North America. I see lots of kids playing near their homes on their own in the neighborhoods I live in. Parents aren't too far away and they're with the other kids of the neighborhood. Maybe the safe distance a kid can travel on their own hasn't changed, it's what you could actually accomplish in that distance that has changed.
Today's society, especially western society, is much less community focused. So now, not only do I not know what strange adults exist in my 200 home open air neighborhood with a single playground, none of the adults there know me either nor my family and therefore a perceived sense of risk and maybe even actual risk is higher that my kids are in danger.
That's what happens when population explodes and population density increases. All that's happening is the West is finally experiencing what many Asian and Middle Eastern Countries already knew.
- Data Scientist MIGHT mean Applied ML Scientist (important distinction) - Data Scientist MIGHT mean Data Engineer - Data Scientist MIGHT mean ML Engineer - Data Scientist MIGHT mean Data Analyst - Data Scientist MIGHT mean Statistician - Data Scientist MIGHT mean Product Analyst
The traditional idea of a data scientist for the last decade or so is someone who is able to do insight extraction, create models (ML and otherwise) and build dashboard and presentations. In practice this has mostly proved to not be practical and little value is being extracted from the role, so a mature organization will properly break out the responsibilities into the above mentioned functional areas.
I think it's super important that you reach out to industry data scientists at companies or in domains you're interested in and ask them what it is that they actually do on the daily. Be careful with most data science roles as they are really just data analyst roles in disguise.
Very few true data "science" roles exist, and I'd argue that might you not actually want to work in those roles since they likely exist in companies that have no idea what they want out of them.
That being said I think the 2 absolutely crucial technical skills to have for almost any modern data related job will be:
- Python + Pandas - SQL
That's really going to be the technical foundation to make a data career in industry.
The breadth of other skills and required knowledge is too much for any single person or post to tell you.
I'm not entirely sure what your PhD is in, but if it has an associated domain in industry and if you're actually interested in that domain, I would recommend starting there and seeing what Data Scientists with similar academic backgrounds as you might be doing. LinkedIn is a great place to find people and connect!
Best of luck.
The reality is this profession isn't that hard, and majority of people working in it are pretty much just plumbers using the innovations of true computer scientists.
We've managed to created a much more inefficient gatekeeping mechanism than just creating a proper certification process and commended ourselves for it and pretend it's somehow more meritocratic than just getting a comp sci/eng degree and license.
I think younger people who want to progress and learn a ton through osmosis, and develop their professional networks should be in person.
I think the future needs to be all about flexibility, with well understood trade-offs.
The companies which offer this flexibility will attract the best talent.
That being said, Web 2.0 had two primary themes and business value propositions. It was social media/ads, and e-commerce. Those were pretty clear and there was no confusion about how money would be made. Out of those was born what I like to call Web 2.5, where basically every industry was getting SaaSified in one form or another. Out of this was born big data and the emergence of the most recent ML hype due to the need to create these algos to maximize ad driven profit.
Crypto/Decentralization/Web 3.0 has a much much more challenging road ahead to get mass adoption. Outside of the obvious scaling issues decentralization brings about, decentralized apps have a much less clear path to monetization. That's in essence why they need to create their own economies with tokens and crypto. Really for "Web 3.0" to be successful, we are looking towards a whole scale reimagining of how economies function. This is entirely different than using the internet to create analogs to real world things like stores or newspapers.
The value proposition of social media was abundant, and the path to profitability was clear. Ditto for ecommerce platforms. Ditto for the majority of SaaS applications.
I think the fundamental divide here is talking about Web 3.0, (or ICOs, DApps, DeFi, NFTs, or w/e the "use case" of the day is), generally seems to be avoiding talking about the true value proposition of decentralization, which will face such tremendous resistance from powerful world governments, that it's akin to a war being fought to creating a new world order and evolution of how human society is shaped. The stakes are much higher and failure is much more likely.
I wouldn't be surprised if all the most successful projects just end up being centralized layers on top of the protocols and all they really end up doing is usurping the place of some existing Web 2.0 platforms due to some network effect pushed forward by a few key influencers. That's essentially what happened with TCP/IP and HTTP. I don't see why it will be any different this time.
It's an opportunity to assess culture fit. Now many will argue and say "no I use that time as an opportunity to let the candidate ask me all the tough questions!". These people are in denial and are following a ritual they've been subjected to, but haven't really thought about it objectively. This process is entirely subject and allows unconscious bias to penetrate the process. When over 85% of the interview is you answering questions, it's an unbalanced situation.
If you truly want to learn if the company is for you, use LinkedIn search and see people who've left the company recently, preferably in the same role as you. The smaller the company, the more effective this will be. Many won't want to speak to you since they might be concerned about employer retaliation, but you can pretty much read the subtext if they are willing to talk and dance around issues.
Keep the question you ask in that 5-10 minutes light and fun. It's primarily being used to assess if they like you. If you feel your interviewer is a very technical person, ask them to gush about all the awesome technical things they've done. If it's a more behavioral interview, make your interviewer feel good about themselves by asking leading questions so they can brag about how great the company is.
- verbal anything is worth less than 0 - Don't be the chaser, be the chased
As I've gotten older and more experienced, I've also started to reject interview processes I don't like or offer feedback if it didn't make sense. Obviously I only do this if the risk is low.
I've thought a lot about this, and I think it has to do with the innate human desire to belong to a tribe. I also think that's why army's will tend to recruit people on the younger side of adulthood, but also why companies that require a lot of hard work and long hours will do the same. Younger adults with less life experience (the under 25s), generally speaking (obviously there are always exceptions), will tend to be the target audience for companies and armies simply because they are much more likely to develop some form of loyalty/tribalism. I also think a part of it is due to the fact that most humans are still developing their worldview at that point, and so if you get them at this stage, you're much more likely to mold them to buy into your vision.
If you want a laptop to actually run AI/ML/RL workloads that would benefit from a graphics card, you're better off optimizing your search for decent specs, no dedicated GPU, portability and good battery life, and spending the money on a powerful desktop setup instead. Your workflow can look like this:
1 - design experiments on laptop or use laptop to remote into desktop with GPU 2 - scale experiments on desktop with GPU till you reach a point where it's maxed out 3 - use Google co-lab with GPU or Paperspace or other similar services (usually offer a free tier for experimentation) 4 - scale on cloud if you really want to parallelize
If you have no choice but to buy only a laptop due to budget, I still think you'd be better off purchasing a laptop without a dedicated GPU and going for points 3/4 above. Seriously, the gains from using an RTX on a laptop are just not worth the cost.
The benefit of using colab or other types of cloud services, is that your investment is low to 0 and will allow you to determine if you actually should invest in your own dedicated hardware.
Also, remember that even having these types of thoughts means we are in a relative place of tremendous privilege where our basic needs are met and secured. Many in the world don't have that privilege. That's not meant to guilt you, but to help you put things in perspective. Usually when we are materially satisfied, we start to seek higher levels of meaning and since we spend so much of our waking time working, we try to seek satisfaction through work. But the world doesn't value our individual satisfaction.
Finally you seem to have some conflicting goals (or maybe not). If you want to retire early and stay with your partner, it might mean making the sacrifice of living in London and also working in a high paying job where you are disconnected from the value it brings to society. You can potentially speed up this process by working on building a high value company, but working in quant finance you should be able to assess risk-adjusted returns and determine if that path is feasible or makes sense in your individual circumstances.
Start ups hiring people to be ICs and pay them like ICs (in the form of equity) that early is a bad signal. They should be hiring contractors instead if they need specific work done as an IC.
The first bunch of full time hires are leadership hires. You will wear lots of hats, and do lots of things you wouldn't do as an IC at a typical large company.
Beyond that, you need to be absolutely sure you will mesh with the "founders". You are in essence a founder. You are likely building something that will be foundational to the company. You will be making decisions that will probably live beyond your tenure at the company.
So either you need full autonomy, or you need to know that conflict will be handled properly and your opinion will weigh just as much as the "founders".
Love this. These are luxury products for upper middle class folks in a lot of developing countries. I don't think people really grasp this until they experience it themselves.
You could also try getting into an analog design company and work on their software stack, but generally that would require embedded development skills at the least.