vikp

I think you're right, if you view "education" as content delivery. I think that for the most motivated people who are willing to learn on their own, education is often just an access problem (hence why so many smart people have focused on creating MOOCs and other content delivery platforms).

But for the people who aren't self-motivated (which I think is the vast majority), education is more of a motivation problem. Many people in this group (including myself) were trained by the traditional educational system to view learning as a chore or a slog.

I've talked to many people who have started MOOCs, but gave up halfway through due to boredom. We often see this as the student's problem/fault, but looking at it that way means that you'll only ever focus on the small percentage of students who are already motivated.

The motivation problem is one that just delivering more content won't fix. I think solving it is incredibly impactful, and can create far more opportunities than the content delivery platforms have. Unfortunately, most solutions to it are expensive (colleges/bootcamps), or lack depth.

I barely made it through college, and worked in factories and warehouses during/after school. Because of my bad grades, I thought manual labor jobs were the only ones I could do.

I later worked overseas (in Guyana). I've seen many bright young people worldwide forced into doing work that crushes their spirits, because they either never had access to educational opportunities, or they struggled with rigid educational systems.

Education is one of the few ways to reduce systemic economic inequality, but it's getting harder to get a cheap, high quality education over time.

I'm building Dataquest, which to start is teaching data science online at low cost. Dataquest teaches skills in depth, so students really understand the concepts and can apply them with projects (vs most syntax focused online learning). We've had students get jobs at companies like SpaceX and Amazon.

I hope the positive impact I'm having is in some small way helping people achieve things they never believed they could.

I'm the founder of Dataquest (https://www.dataquest.io) -- we teach data science online from the basics, and have a comprehensive curriculum that includes machine learning, spark, and data visualization. You can skip the Python basics and start with more intermediate/advanced material (and build your own projects!).

We also have a data engineering path that teaches more CS fundamentals, and may be a good fit (this is still being developed, but has a few courses).

In the original blog post announcing the Gitlab acquisition (http://blog.gitter.im/2017/03/15/gitter-gitlab-acquisition/), there's mention of open sourcing the mobile apps -- "Next piece of wow: we will be open sourcing all of the Gitter. That’s right, the web application, the mobile apps, the whole nine yards, free and open."

I've worked across a variety of industries, from warehouses to factories to big government bureaucracies to startups. I'd argue that the problem isn't that people don't have the curiosity to become lifelong learners. Almost everyone is enthusiastic about something (most people have hobbies they spend their own time on). The problem is that their workplace (and to some extent, society) don't reward curiosity.

If you're a programmer who works at startups, it's easy to think of learning as part of the job -- your tools and conditions are constantly shifting, which makes learning part of the work culture.

Unfortunately, most industries don't work that way. At a warehouse, you do the same thing from the first day you start to the day you leave. It's much the same in the government and at other large bureaucracies.

The problem here isn't with our educational system, it's with the modes in which we work. And to be honest, that's fine. Most people who work in industries that have static work conditions optimize for other things, like family time. It's also very hard to see outside of your silo and imagine a future where you're doing different work.

Most people are optimizing their life for right now, not their theoretical career 10 years out. Unfortunately, there are fewer stable lifelong factory jobs every year. This forces people to move outside their comfort zone, which makes lifelong learning necessary.

I don't have an answer to the problem, but some potential solutions are:

* Reward employees for taking time outside of work to learn skills

* Show people examples of others like them who have learned for motivation

* Find something a person is really interested in (not because it's a "valuable skill", but just because they think it's cool), and help them take small steps to start learning it

Basically, I think it's possible to help people transition to new careers further into their lives than you do. I don't think it's ever too late for someone to learn a new skill, if they have the right incentives.

Dataquest | Data Science/Data Engineering Instructor | San Francisco | Remote okay | $90k-$120k + equity

At Dataquest (www.dataquest.io), we teach data science and data engineering to thousands of students around the world every day. We teach the concepts behind the code, then help students build projects until everything clicks. We get results -- we've had students get jobs at companies like SpaceX, and our NPS is around 60.

We don't believe in skimming the surface of concepts so that students just know what to type. We build intuition around complicated ideas like random forests from the ground up. We walk students through building their own algorithms, so we can help them understand the tradeoffs and limitations of techniques. We help students build projects, so they can solidify their knowledge and get real-world experience.

We're looking for someone to help us extend and enhance our curriculum. This involves writing instructional content, creating projects, thinking about how to improve how we teach, mentoring students directly, and getting feedback from students to improve our teaching methods.

We're looking for someone who's passionate about teaching, and shares our mission to give people access to high quality education at a low cost. You'll have a lot of latitude to shape how we teach, and to help build a personalized educational platform. We're bootstrapped and profitable, so this is also a chance to learn more about the business side.

This is a great role if you want a make huge impact on the world, grow a business, and rapidly expand your skillset.

If this is interesting, please email vik at dataquest.io.

Dataquest | Lead Software Engineer | San Francisco | Remote okay | $100k-$140k + significant equity

At Dataquest (www.dataquest.io), we data science to thousands of students around the world every day. We teach the concepts behind the code, then help students build projects until everything clicks. We get results -- we've had students get jobs at companies like SpaceX, and our NPS is around 60.

Our infrastructure is built around React and Python 3, and we deploy using Docker and Kubernetes. We interactively run and check student code, a major technical challenge.

We're looking for someone to own our technical stack and direction, while also leading and growing our engineering team. Ideally you've built something at scale, and can balance adding customer value (shipping fast) with clean code and architecture.

We're looking for someone who's passionate about teaching, which both involves sharing our mission (to give people access to high quality education at a low cost), and wanting to mentor other engineers.

You'll have a lot of latitude over technical decisions, and you'll have significant input on product decisions as well. We're bootstrapped and profitable, so this is also a chance to learn more about the business side.

This is a great role if you want a make huge impact on the world, grow a business, and rapidly expand your skillset.

If this is interesting, please email vik at dataquest.io.

8% isn't a bad click through rate, especially if you're sending newsletters weekly. I don't see anything about frequency next to the subscribe button on your homepage, so people may have been expecting something different, which would drive down the click through rate.

I've also seen enlight.ml posted here before, so my guess is quite a few of the subscribers came from HN or sites like it. Typically the users you get through these kinds of postings aren't all in your target audience. This might also be resulting in a lower click through rate.

The key to getting a high click through rate is delivering a lot of value. There aren't any good shortcuts.

Here are some things that may help you figure out and boost your perceived value:

* Send a survey to your users asking them what types of updates they want to get from you, and how often.

* Figure out if the same 8% engages each time, and what the patterns of click throughs are (do some people click through every single time?).

* See if anyone on your list would be willing to chat with you about the value they get from your newsletter/site.

* Do some content marketing/promotion to attract a more targeted audience.

* Make sure you're selling the benefits of doing the lesson in your emails.

* Have some clearer messaging around the signup box for the list.

I run a site to teach people data science, and I've learned a lot of marketing lessons along the way. Happy to chat further if you want -- my email is in my profile.

Dataquest | San Francisco, CA | Data Scientist| Full-time | Onsite or Remote

At Dataquest (https://www.dataquest.io), we're building the future of online education. Most online learning has turned into "let me throw up a bunch of videos and call it a course". We help people actually learn and demonstrate their skills . Our paths take people from zero coding knowledge all the way through to getting jobs at companies like 3M and Fitbit. Our cohesive curriculum, interactive learning by doing approach, project focus, and strong learning community are critical to this process.

We're all self-taught, and we're passionate about helping people learn more effectively. We're constantly trying new learning techniques and refining our approach. We're profitable and control our own destiny, so we have a lot of latitude to do what's best for our students, not what's best for VCs.

We're been teaching data analysis and data science to tens of thousands of students a month, and we want to deepen our curriculum. This involves teaching more complex math and machine learning techniques, how to more effectively communicate results with interactive dashboards, and much more. Our philosophy is to teach not just syntax, but the deeper understanding and theory behind techniques. Our lessons are usually walkthroughs of interesting datasets, where we introduce concepts along the way. Looking at some of our blog posts (https://www.dataquest.io/blog/data-science-portfolio-project..., https://www.dataquest.io/blog/kaggle-tutorial/), and our lessons (https://www.dataquest.io/track/data-scientist-track) might help explain our approach.

This position will have a lot of autonomy in expanding and building our data science curriculum. If you're passionate about finding insights in data and teaching others to do the same, we'd love to chat.

Email me at vik at dataquest.io if this sounds interesting!

I wouldn't pay for daily news, which is usually reasonably binary.

I would pay 10-20 a month for a site that had high quality in depth analysis and articles, though. Something like a paid version of longform.org. The articles wouldn't have to be tied to current events at all.

Dataquest | San Francisco, CA | Data Engineer | Full-time | Onsite or Remote

At Dataquest (https://www.dataquest.io), we're building the future of online education. Most online learning has turned into "let me throw up a bunch of videos and call it a course". We help people actually learn and demonstrate their skills (we're all self-taught). Our paths take people from zero coding knowledge all the way through to getting jobs at companies like 3M and Fitbit. Our cohesive curriculum, interactive learning by doing approach, project focus, and strong learning community are critical to this process.

We're constantly trying new learning techniques and refining our approach. We're profitable and control our own destiny, so we have a lot of latitude to do what's best for our students, not what's best for VCs.

We've been teaching data analysis and data science to tens of thousands of students a month, but we want to expand into teaching data engineering. Because we teach interactively, this both involves building backends (like Spark clusters), and teaching students how to use those backends in end-to-end data pipelines.

This position will have a lot of autonomy in creating and building a data engineering curriculum. Nobody is teaching data engineering well online, and this is a great way to shape the future of the field. If you're passionate about building data infrastructure and teaching others to do the same, we'd love to chat. We'd prefer strong knowledge in Python, Docker, and PostgreSQL at the minimum.

Email me at vik at dataquest.io if this sounds interesting!

It's interactive in the sense that if you download the notebooks you can work through them with a local Jupyter installation.

This is very specific to data science for now, but I'm working on Dataquest (https://www.dataquest.io), which has interactive explanations and diagrams. It's almost entirely reading theory and explanations + writing code. It's hard to get broad coverage like udemy/coursera, but it's a problem we'd love to work on in the future.

I've liked scotch.io for some dev things, and thinkster.io for some Angular/frontend.

Also check out these IPython notebooks for a ton of topics: https://github.com/ipython/ipython/wiki/A-gallery-of-interes... .

Mark Zuckerberg donated 100 million dollars to Newark public schools a few years ago, and they ended up no better for it (and arguably worse off). Here's a great article on the topic -- http://www.newyorker.com/magazine/2014/05/19/schooled .

The main reason for the failure of the donation to create positive change was that the community wasn't engaged at all in the reform process. The first time Newark students and parents heard about the donation and the accompanying "reforms" were when Zuckerberg announced them on Oprah.

I completely agree with your sentiment, and wish philanthropy wasn't seen as an unmitigated good thing.

Dataquest.io | San Francisco | https://www.dataquest.io | Onsite / Remote

We're an early stage startup that teaches data science interactively in the browser. You learn by analyzing data and writing code; not by watching videos.

We're all self-taught and passionate about helping others do the same. Our revenue is growing quickly, and we're looking for a couple of people to join our team:

Data Scientist -- create lessons on topics like Spark and random forests that will be used by thousands of students. Create innovative learning methods (interactive coding exercises, visualizations, etc). Python and R experience preferred.

Frontend engineer -- redesign our learning interface. You'll have huge creative freedom to build things like realtime collaboration, hinting, and more. Design and angular experience preferred.

Email vik@dataquest.io if you want more details.

Off topic, but I use Jekyll for my blog, too.

It's compiled to static HTML, sure, but the templating is pretty powerful (downsides of Liquid aside). It's easy to make global changes with templates and CSS.

Octopress can make a lot of Jekyll tasks easier -- https://github.com/octopress/octopress.

Building a way to learn data science online, using python -- https://www.dataquest.io

Yes -- I write ~95% of my python code in 3, and scrapy is the only library I commonly use that doesn't support it. Here's a listing of package compatibility: http://py3readiness.org/.

I went from being a US diplomat with no coding background to getting a job at edX as a machine learning engineer, so it's very possible. The keys are to find projects and build a portfolio so that you can prove your capabilities, and to start a blog/go to meetups so that you can build an audience and find opportunities.

There's a project called Tulalens (www.tulalens.org) that could use help making an android app.

They're building "yelp for low income people in the developing world". They surveyed pregnant women in low income areas around Hyderabad (India), found what hospitals they went to, and asked them rate their experiences. Some women were having to pay bribes to get seen, and others reported a low quality of care. Tulalens was then able to compile the data, and go back to the women and tell them the best hospitals in the area. Many women switched as a result.

For now, they want to build an app to scale up this process, and give results to users in real time. There is some open source code already.

If you're interested, email priya@tulalens.org.

Disqus is good for an MVP. It's easy to embed and works well.

If you're worried about "owning" the data/community, you could host your own discourse (http://www.discourse.org/) instance and use their embed feature (http://eviltrout.com/2014/01/22/embedding-discourse.html).

Embedding is sadly missing inline commenting, though.

It depends on how much data you can include in the prediction phase. If you train with karma earned right at the time of the article submission, and then predict with the exact same methodology, it will work fine. If you put in a "fake" karma value when predicting, then it will of course not work well.

Dataquest (www.dataquest.io) | Boston, MA | Software engineer, data scientist

Dataquest is an early stage online education company making "codecademy for data science". Since we launched a few months ago, we've taught tens of thousands of students around the world everything from basic python to data visualization to machine learning. We're expanding, and looking for a full-stack developer and a data scientist to help people learn, apply, and prove their data science skills.

Our tech stack is Python and Javascript (Django + Angular), with docker used for code sandboxing and deployment/testing. A software engineer would focus on building and improving the learning platform, whereas a data scientist would focus on content, community, and analytics.

The salary range is 75k-90k, with significant equity (2-5%). This is a great chance to have a huge impact on the development of an early stage startup. We have an office in downtown Boston, and great investors.

Email vik@dataquest.io if you're interested.

Great change.

I second the request for examples -- "gratuitous" is subjective, and until better moderation tools exist, it would be nice if the community has some standards around who to issue "gentle reminders" to.

It would also be nice to have the guidelines be in a more prominent spot than the bottom of the page -- it's hard to notice them as is.

You can actually create edX classes with existing resources. The xblock spec is pretty flexible, and the existing blocks can be used with a lot of different types of content.

There are also a ton of companies that host open edX instances, which let you make your own courses. Several offer free trials -- here's a list: https://github.com/edx/edx-platform/wiki/List-of-Open-edX-se... . There's a nice open source ecosystem forming. I know at one point, edX was going to open edge.edx.org and let anyone make courses, but don't know where that's at now.

Your platform has differentiators from edX, but these aren't necessarily them.

Nice looking site! I worked on something similar a while ago, but it didn't look nearly as good as this.

The problem I faced was in attracting users without one key compelling subject area. Saying "this is a platform for every type of class" is awesome in theory, but it's hard to find people who are just browsing around for random classes.

I'd love to be able to browse at least one complete course without having to login, also.

On a sidenote, livereload is still linked in your javascripts, and is hitting localhost. You might also want to concatenate and minify the js.

See this issue: https://github.com/jtleek/datasharing/issues/217 .

I'm also curious about the general case of people believing that we're in a bubble giving advice to startups. Do they advise that startups keep a really low burn rate and wait it out, or go for broke and raise as much as possible now? Do they advocate certain business models that do better in recessions?

When I applied, you took the test. If you passed the test, you then had to go online and submit ~10 short essays. If you passed that, you went on to the oral assessments. There may have been a resume component involved when you submitted the short essays.

Thanks for sharing! Interesting how we went through a lot of similar steps. I'm glad things worked out well.