a small thread on self-learning, CBC, reallyconfused and what comes after. i tried building reallyconfused to solve the problem of self-learning. my aim with the product was that i could help someone just like me navigate the field of tech easily through curated roadmaps.
so what's the consensus? self-learning is… kind of a… hard problem to solve. there's no one fits all solution. people who are sufficiently motivated to learn by themselves are not an ideal paying customer because they could use the free resources on the internet to learn by themself.
and if i was indeed the target customer, i would've not paid for this service. i got the distribution and virality part of it right, the roadmap format was highly effective too, but i really did not figure out how to monetize it. this was what YC was concerned about too. so what now?
during the peak of reallyconfused's launch, @adammikulasev reached out to me through reddit "Just saw your road-map post learnprogramming… wow! I think that is really going to help some of my clients." and we proceeded to connect through linkedin.
after running some experiments with reallyconfused, reddit and discord, we had a successful funnel where we were able to convert a bunch of self-learners into a discord server. small win. what do we do next? how do we keep these motivated people who wanna learn engaged and provide value?
after talking with our users, the common theme was
- hard to get valid real-world work experience
- hard to upskill from tutorial hell to real-world projects
- hiring market is hard for new-folks without real-world experience
we had a light-bulb moment "open-source is the way". it absolutely ticks all the boxes. we use open-source to give people valid work-experience, and leverage that for employment. we do it in a way genuine value is created for all the parties involved.
GSoC did it, and it was effective. @jjmachan comes in and validates the whole thing because because it was the same path that he's taking. only one way to test it out. dive right in!!! and @fastprogrammer was born.
we quickly prototyped a roadmap and set a date for our first free cohort. since all of us had considerable experience with open-source, it put us in the place where we were able to deliver value and we had a mechanism in place to do it.
we had the perfect distribution, a meaningful vision, a group of motivated learners and it was time to launch. we launched the open-source in 7 weeks challenge in our discord and 45 people signed-up for the challenge from our 330 people discord channel.
the goal was to do meaningful open-source contribution, leverage that using different ways, get acquainted with the community and do networking, and use that to get a job. the perfect path for a lost, intermediate programmer, looking to break in to tech.
we had to keep the incentives right because we did not wanna create more trouble for the maintainers like hacktoberfest did, and wanted to create genuine value for the open-source community.
we specifically focused on closing existing issues and incentivized solving meaningful bug-fixes and features. we wanted folks to stick to one project to make impact than jumping around.
it worked. our roadmap, with dedicated channels, stand-ups and calls were super effective. our community was active and filled with awesome people, and we were able to produce kick-ass, tangible, valuable results.
- closed 18 issues
- fixed 8 bugs
- added 10 features
- added 3 features that are awaiting approval
- had 27 pull requests merged
- contributed 30 pull requests
we weren't satisfied with this result. we wanted better results for our participants. we incentivized people to tackle challenging feature and added three more steps to the journey. in a world where half-assed personal projects are used to spam recruiters through linkedin, we tried to incentivize meaningful contribution.
we tried to get all our participants to show-case their contribution to actual, useful open-source projects and helped them polish their resume. we used our combined knowledge to curate strategies on how you could network, but without using linkedin, without pretentiousness and by showcasing actual value.
it was a spectacular success. JJ after contributing to BentoML, got an awesome paid internship at BentoML itself. along with that, two of our other participants got internships and jobs through this program.
it's one thing to have success with an open-source challenge but whole another ball-game when people get jobs through the roadmap that we designed.
so what worked for us?
- supportive community
- tangible outcome
- resources and the network
- accountability
- mission
why this was a better model than reallyconfused self-learning? first and foremost, we focused on the user-journey and the community than the product. we don't even have a landing page. the people who joined us were the ones who was committed to the mission of upskilling themselves.
there's too many resources out there for beginners, but hardly any good ones for the people who got plateaued while learning to code. stuck on projects, or stuck taking too many courses? how do you get that valuable real-world experience that's so needed to land a job.
i think we managed to tackle that niche, and tackle it well. so what are we doing in the future? double down on the current model. personally, it was a great, fulfilling journey to take part in this because of the outcome, the process, and getting to work on something meaningful.
if you want access to our high quality resources, community, mentor resource and other fun things, join us for the next cohort. if you're interested, sign-up here https://forms.gle/jExh2CAXMJq3ys2i7 sorry, we can't afford typeform, yet.
we don't really have a landing page, but you can follow us at @fastprogrammer for other cool updates regarding our journey. if you want me to dig into any other aspect of this particular experience, let me know!