Memorang (https://www.memorangapp.com) | Los Angeles, CA | Full-Time | Remote-OK | Software Engineering Lead |
# Overview
We're building the world's first content-agnostic study platform and exam marketplace from grade school through graduate school. Our mission is to automate and innovate upon cognitive techniques and strategies to level the playing field in education, especially as it pertains to knowledge acquisition, retention, and mastery.
The core of the platform is our data model that supports complex content, from art history to engineering certifications (think Google Knowledge Graph). By unifying the crowdsourced community and premium content marketplace (for 3rd-party businesses) we aim to turn Memorang into a lifelong learning platform for hundreds of career paths. Think of the product as an education-focused chimera of Anki, Kaplan, Dropbox, Amazon, and Squarespace.
# Traction
200K monthly learners (20X growth in 2 years), 30% of US med students, profitability (good unit economics), just raised $500K funding.
# Opportunity
This is right up your alley if you:
- Want to own the tech stack and roadmap
- Want to lead/grow a team (i.e. tech lead/CTO)
- Geek out on spaced repetition, cognitive science, and machine learning
- Like making immediate, positive impacts on users that empowers their future
I can completely relate to that frustration. I came from an engineering background before I went to medical school and I struggled a lot with the different styles of learning (applied principles at MIT vs. knowledge acquisition in med school). The biggest mistake you can make is think that your ability to remember is just an innate ability instead of a skill that can be cultivated.
The problem with spaced repetition software is that "SR" is just one of many known cognitive principles to aid in learning and memory. The encoding process is just as, if not more important than, the review process. I highly encourage looking into interleaving vs blocking, overlearning, learning vs. performance, retrieval-induced forgetting, contextual variability, and the (pre)testing effect. Learning how to incorporate these into your existing study process could really help you out.
This is where great existing software (e.g. Anki) falls flat for many people. If you're interested in seeing an alternative approach you can check out my company's platform Memorang (https://www.memorangapp.com). We recently raised a $500K seed round and are working on some really interesting problems around learning and memory, including a new type of spaced repetition that deviates significantly from the deterministic algorithms like supermemo3/Anki. Probably even more interesting, in addition to building apps for students we're also implementing a platform for developers and education entrepreneurs to monetize their efforts.
Although we're still fairly new we finished second to duolingo at the international reimagine education awards last week. Check out Memorang (web/Android) and let us know what you think (or come join us as a dev --> [email protected]).
Currently you can export the HTML and also as printed sheets. In the next few months we'll enable exporting to formats such as csv/tsv. (This is currently supported in the admin interface, but we haven't made it a user-facing tool yet).
If it's any consolation, we already prepaid the next 2 years for AWS and are cash-flow positive. (i.e. not going bust anytime soon).
Thank you for the questions. I think it's likely multi-factorial and to be honest there are many features we haven't yet built that Anki has and are in our backlog... there's still much work to be done. It's likely some combination of expert content, multiple learning modes, gamification, friendly UI, and collaborative features. Give or take the importance of one or the other depending on the person. Some struggle with time management, others study fatigue, and others just don't like or trust user-generated content. For my own studying, I'm not a big fan of pure flashcards and prefer other methods of active recall.
I'm not sure what you are angry about, but please let me try and clarify a few things:
>A better app that doesn't crash as much
Working on it. Worth noting that it's free to download.
>Thousands of free decks
Memorang has about 5 million free flashcards currently as part of the crowdsourced ecosystem. You can get content in 4 ways: creating, importing, searching, and purchasing. You can import from Quizlet, Excel, Anki, StudyBlue, and Cram, or collaborate in groups on common topics.
>An open ecosystem
While I think we could do a better job with content discovery, the collaboration tools on Memorang do not exist on Anki. If you wanted to work together on an Anki deck, it would have to be with some combination of dropbox or git. My opinion is that Anki is pretty antisocial.
>Source code
I think HN readers know more than anyone that there's nothing wrong with not open sourcing your company's code. Developers need to eat too. That being said, we are open to making our API publicly available to build learning applications on top of.
>Doesn't spam your email
I welcome your suggestions regarding email and onboarding. Currently we send a welcome email and account verification. Over the course of 1-2 weeks we give short snippets of how to use certain learning modes on the platform. You can unsubscribe from emails at any time via the unsubscribe link or via your account settings.
>Isn't an obvious monetisation of anki's general premise
It's worth pointing out that the creator of Anki has made tens of millions of dollars off of charging for downloads. I'm not sure what "premise" you are referring to - but ours is that the core learning application should be free and not a paid download.
>How anyone thinks that paywalling a bunch of flashcards is a good idea (when anki offers them free) is beyond me
Crowdsourced content is free on both Anki and Memorang, so I think you may have missed that the "paywalled" flashcards are actually an optional upgrade. As a bit of background, we've spent thousands of dollars on content creation and are committed to continuously upgrading and maintaining that content. For example, our USMLE Step 1 and MCAT content are one-time purchases for lifetime access and we maintain updates to stay current with the latest medical information at no extra charge. Users are happy to pay for that level of trust and security.
I hope the above information helps you better understand what we are doing.
> Could you please create a mailing list for different professions/subject areas so potential customers could be alerted when relevant premium study sets are released?
That is a fantastic idea, thank you so much! We can definitely create a few mailing lists and I'd be happy to keep you all in the loop. We currently have 10 new subjects in development and are scaling to release at a rate of (ideally) 3-4 per month by next summer. We hired a director of content specifically to recruit teams of experts and to oversee this creation process more fluidly. I'm going to set this up ASAP.
Send your email to [email protected] and your intended subject. It's possible one of our next releases will be appropriate for you.
Thank you, it really means a lot since we've been working on this for over 3 years now (finally hiring employees this past month!). I wish the best of success to you too.
In reference to people wanting freebies - that's just human behavior. If they see sufficient value in a product (e.g. iPhones), then they will happily pay for it.
We're starting with the assumption that learners will not be able to religiously follow a schedule. That means that whether you're cramming, checking in sporadically, or completing a daily queue - the algorithms should ideally adapt accordingly without making your "to do" list untenable. One component of this is determining whether your long-term retention is bounded to a specific timeframe (e.g. learn content X by future date Y) or unbounded (lifelong learning).
What we're borrowing is the concept of memory decay, but I don't think there are many similarities beyond that. Imagine learning spanish in college: you could set a goal for your final exam but also have a lifelong learning goal. Whether you're following your schedule or cramming, your answer events will communicate with the intersecting algorithms so that you're being optimized for several different use cases simultaneously. When we launch the "goal setting" feature, gjcourt may write a blog post about it and post it to HN...
It's an interesting thought! Boredom can definitely lead to study fatigue, but the people in this example were medical students preparing for a high-stakes board exam at a top-10 institution. Boredom might be less of an issue at this level - it's more about efficiency and a tremendous feeling of stress and pressure. Indeed, several of these students who were in danger of failing were studying as much as 10-12 hours per day... essentially spinning their wheels in the mud.
That being said, combatting boredom is a really important thing to consider.
You can also click the search bar in the top left and go through the user-generated database. I believe there are about 5 million flashcards there currently.
I see that you're angry, but I think it's because you misread my comment.
When I wrote "I encourage you to learn a subject towards 100% mastery and you'll see exactly what we mean" it was purely meant to encourage you to learn a small subject on Memorang to help answer your questions about the methods you inquired regarding interacting with the learning system. It wasn't, as you seem to believe, a statement that you couldn't learn via alternative methodologies.
We have this at the admin level and plan on making this a user-accessible feature as soon as we can.
>Does your service offer benefits to people who aren't and don't have the resulting new card review burden or need for cramming?
It depends on your goal. In its current iteration, Memorang functions very well for mastering new concepts on a shorter-term scale (e.g. hours, days, or a semester). However, long-term retention on the order of years is something that we are working on a different implementation of. That will be a huge focus in the next semester after we finish shipping this 2.0 launch in January.
>Can you elaborate more on what exactly memorang does that keeps people "engaged in vivo"
Between algorithm choices, learning modes, and other customization options there are dozens of ways you can learn the same material instead of simply flashcards like Anki.
For the other components, some of these are works in progress and others are already present. I would encourage you to learn a subject towards 100% and you'll see exactly what we mean. Part of the answer is gamification via scores and leaderboards, some is advanced study stats where you can compare your progress, gain insights into how you learn, and see how you stack against the global average. Another part of the answer is targeted feedback via the "tutor" which will introduce study tips and humor depending on your progress.
We also are launching a 2.0 version very early next year which will be a fruitful step in the right direction.
Overall, we have a pretty broad vision for what we think we can accomplish and are actively working towards getting there from a tech point of view (development) and via academic research with well known universities. The ultimate goal would be that you could just sit down, select what you need to know, and then be perfectly guided towards subject mastery in whatever way is most optimized for you as an individual.
Don't have strong opinions one way or the other on Knewton. That being said, I generally feel that enterprise software where you sell to institutions and publishers is fundamentally disconnected from software that's targeted at the users directly.
I would greatly prefer to hear students raving about a learning product rather than their instructors. In your case you seem to be a bit underwhelmed by what they offer.
This message really resonated with me. Thank you very much for sharing. One thing that we've really thought about is how to learn in the real world for subjects outside of school. I know this is a silly example, but check out this set:
http://www.memorangapp.com/flashcards/20746/How+to+cook+the+...
We're working on a version of our platform where you could define need-to-know facts for your employees and then give them an engaging, adaptive way to learn. These data would be fed back to you and you could determine how to help your employees, or who is falling behind. Maybe even aspects of your "curriculum" for employee training that are insufficient.
I can't answer with an exact number until we can do sufficient prospective studies, but I can share some data: We have done some studies where two students were already using Anki and were failing or in the danger zone of failing. After being introduced to Memorang for 6-8 weeks, they all increased their scores by a national standard deviation and had a 100% passing rate.
What does this mean? The students said that Anki didn't motivate them, and that study fatigue (Not the spacing effect) was the key differentiator. As a medical student I am very familiar with this, and so one of our main goals with Memorang is to keep people engaged in vivo, not just have an optimal algorithm should you religiously follow it to a T.
What you're describing is exactly in line with what I've experienced anecdotally and what we've seen in our studies and retrospective analysis of tens of millions of questions answered.
I think the problem with Anki can be distilled down to root assumptions it makes about how people should ideally learn and how to model what is forgotten over time. In studying for a course there are basically three strategies: (1) cram at the end (2) study reasonably throughout the whole course (3) try to keep re-learning the course indefinitely.
We've thought long and hard about how we can cater to these different styles and think we've done a good job with a hybrid of #1 and 2. What would make more sense is something coming early next year as an option feature, which is inputting the date of the final exam of your course and then periodically adding in flashcards to a growing playlist that optimizes retention for the day of your exam. After you finish, you don't necessarily have to continue, but you could set a longer-retention date. An example of this would be that you have weekly quizzes and a final exam, and the algorithm could adjust the date for each quiz while balancing the overall arc towards the final.
# Overview We're building the world's first content-agnostic study platform and exam marketplace from grade school through graduate school. Our mission is to automate and innovate upon cognitive techniques and strategies to level the playing field in education, especially as it pertains to knowledge acquisition, retention, and mastery.
The core of the platform is our data model that supports complex content, from art history to engineering certifications (think Google Knowledge Graph). By unifying the crowdsourced community and premium content marketplace (for 3rd-party businesses) we aim to turn Memorang into a lifelong learning platform for hundreds of career paths. Think of the product as an education-focused chimera of Anki, Kaplan, Dropbox, Amazon, and Squarespace.
# Traction 200K monthly learners (20X growth in 2 years), 30% of US med students, profitability (good unit economics), just raised $500K funding.
# Opportunity This is right up your alley if you: - Want to own the tech stack and roadmap - Want to lead/grow a team (i.e. tech lead/CTO) - Geek out on spaced repetition, cognitive science, and machine learning - Like making immediate, positive impacts on users that empowers their future
# Stack Python / Django, Node, Redis, PostgreSQL, React(JS/Native)
I'm the founder and can't wait to hear from you: [email protected]