Learning Without Limits: How Indigenous Tribes Prepared Me To Master Data Science

AJ Goldstein
Published in
13 min readOct 6, 2017


This is the first in a series of posts on applying Tim Ferriss’ accelerated learning framework to Data Science. My goal is to become a world-class (top 5%) Data Scientist in < 6 months, while open-sourcing everything I find and learn along the way. Here’s the story behind the journey and an invitation to follow along:


There I was, ten yards out, staring my dinner in the face. The only problem was, the wild boar was still alive.

For the past 4 weeks I had been backpacking solo throughout Southeast Asia, and just yesterday had decided to spend the final 2 days of my trip doing jungle survival training in Bario, Malaysia.

With the help of a local guide, I was effectively learning how to live off the land: boil my own water, kill my own food, build my own shelter, and more.

Now here I was, in the most isolated region of the country, face-to-face with my next meal; having to come to terms with yet another deep-seated fear of mine.

Living off the land

In many ways, this was nothing new. I had spent the majority of the past month living with indigenous tribes — finding myself in a long list of situations that had me totally outside my comfort zone.

In other ways, this situation felt just as scary as the last one. And the one before that. And the one before that. Even after facing hundreds of these “oh shit” moments, the fear never went away.

Crossing a makeshift bamboo-tree-bridge over a rapid river just 30 minutes earlier was no less terrifying than this moment of staring a 130-pound wild animal right in the face.

Throughout my 30 days in Cambodia and Borneo-Malaysia, the fear never faded. But, what did change, was how I learned to handle the fear.

Me, my guide Phillip, and the “bridge” that had me so scared I was physically shaking.

Ground Zero

Adjusting to such a different way of life was not always quick or easy.

As part of adopting indigenous culture for 30 days, I sometimes found myself eating insects & reptiles, sleeping on wooden boards, taking ice-cold showers, and using hole-in-the-ground toilets.

Virtually every guarantee and absolute of my life back home had been either stripped away or disproved. Left with only the bare essentials, I quickly found myself at ground zero.

One week before Bario, I was in another village 300 miles south, staying with the Iban indigenous people in Kanowit, Malaysia. I was the first young white male that the children had ever seen in-person and, like every other tribe I’d visited, only a few people spoke more than a little broken English.

The Iban indigenous children & my host-family

As someone who places such a high value on building connection through conversation, the language-barrier was especially difficult for me. I often found myself feeling incredibly lonely and in need of some intellectual stimulation.

Thankfully, a couple weeks earlier I discovered that listening to podcasts was an amazing way to keep myself company. In particular, I’d quickly become obsessed with The Tim Ferriss Show. Each episode, Tim would deconstruct the habits and routines of world-class performers in order to distill actionable tips/tricks for his listeners.

Listening while experimenting with a totally different way of life, I found these podcasts to be the perfect recipe for re-examining my life from the ground up. The combination of fresh ideas with new scenery channeled a level of creativity within me that I never knew was there.

Yum, lunch :P

Accelerated Learning

One theme that really resonated was Tim’s framework for accelerated learning. As a self-proclaimed human guinea pig, he had spent the previous decade mastering various skills like Brazilian jiu-jitsu, language learning, tango dancing, and swimming. In each case, he had become a world-champion in less than 6 months, all by following the same framework (described in detail below).

Hearing his personal stories of emulating the world’s fastest learners completely opened my mind to the sheer breadth of possibility around self-education. In particular, his accelerated learning framework had me asking two questions of myself:

  1. What skill do I want to learn the most?
  2. What fears are stopping me?

The answer to the first question felt somewhat obvious. In high school I’d developed a fascination for data analytics through my childhood love of baseball. And in just a couple weeks I’d be starting my senior year at the University of Michigan — where I’d spent the past three years studying for a degree in Data Science Engineering. So mastering the technical expertise of a Data Scientist seemed like the clear choice.

“What’s Data Science?” Drew Conway’s famous 2010 venn-diagram depicts it as the intersection of hacking skills, math & statistics knowledge, and substantive expertise.

But as I started to try and answer the second question, I came up totally short. Reflecting on the past few summers — where I had three internships involving data analytics — I realized that I’ve continuously shied away from hard-and-fast engineering work, choosing instead to focus in on “softer” business-development skills where I was already most comfortable.

As a result, in none of these experiences did I take things to the full extent of my technical capabilities. Every time, I felt myself just barely scratching the surface of something I’d been seemingly fascinated by for as long as I could remember.

It just didn’t make any sense. Clearly I’ve always wanted to learn this stuff, but why haven’t I given myself the chance?

Limiting Beliefs

On August 17 — my last evening with the Iban people — I was relaxing on the back porch of the tribal chief’s home, listening to yet another Tim Ferriss podcast. This one was a conversation with Tara Brach, a world-renowned meditation teacher.

Toward the end of the episode, while discussing how she’s overcome fear, Tara posed a simple question to listeners that left me dead in my tracks:

“What are you believing that’s limiting you?”

For the past few weeks I’d been forced to come to terms with physical fears big & small, but only now did I begin to consider the mental fears that had been limiting me back home.

Returning back to the second, lingering question above, the answers came pouring out. I quickly grabbed my journal and scribbled down the following entry:

The journal entry from the Iban village, answering Tara Brach’s question

What came out was both surprising and enlightening. By seeing my fears as just that — fears — I was able to take a step back and ask myself if any of these were actually worth being afraid of.

I was able to see that the only blockades preventing me from growth were internal insecurities around my own worthiness and capacity to learn. With nearly all the information I need free and readily available online, the only thing really standing in my way were mental barriers I had created for myself.

Accelerating My Learning

With all limiting walls identified, I was able to begin to knock them down.

Now circling back to Tim’s learning framework, I revisited podcasts, articles, & videos about his personal story, in search for ways to apply the same principles to Data Science.

And by the way, this was possible due to the fact that — in the indigenous villages of Cambodia and Borneo-Malaysia — free Wi-Fi is more available than clean drinking water. Go figure.

So throughout the final two weeks of my backpacking trip — while motorbiking across towns, trekking through jungles, climbing mountains, and riding long-bus rides — I slowly created a learning plan to execute upon returning home.

Below, I’ve outlined an overview of that plan, step-by-step. Please note that the framework described below was originally published in Tim Ferriss’ epic book: The Four Hour Chef.

The Framework

DiSSS: the recipe to becoming world class in anything in less than 6 months

  1. Deconstruction
  2. Selection
  3. Sequencing
  4. Stakes

Step One: Deconstruction

The first step of Tim’s framework is to break down the complex skill you want to learn into it’s simplest parts. The key question here is:

“What are the LEGO blocks (e.g. micro-skills) that make up the big scary wall?”

Two main tools (and supporting examples) for accomplishing this are:

  1. Reducing: break down each micro-skill into its individual components.
  • While learning Japanese, Tim broke up each alphabetical character into native “strokes” called radicals. With only 214 traditional radicals in the language, this turned a near-impossible task — learning 1,945 characters — into something much more manageable.

2. Interviewing: consult experts about learning strategies, key principles, common mistakes, etc.

  • While learning basketball, Tim cold-emailed Rick Torbett (who coached the Warriors to the highest 3-point shooting percentage in NBA history) for learning strategies like “framing the goal on the follow-through” and key principles like “legs for distance, arms for aim” — in exchange for a feature on his blog.

During my first two weeks back home, I started with “reducing” by spending 50+ hours reading every article I could find online about the core/top/essential/critical “skills of a Data Scientist”.

Not surprisingly, much of what I found at first was filled with buzz-words and fluffy terminology, but about 10 articles in, I started to notice some consistent, substantive patterns.

A summary of my first 50 hours of research. In my next post I’ll walk through these findings, step-by-step

Next, I created a list of questions and started reaching out to every expert Data Scientist I could find (e.g. co-workers, university professors, industry professionals, etc).

In most cases, they were more than happy to help. So over the past two weeks, I’ve conducted 10 informational-interviews with local Data Scientists, and learned a ton of “do’s” and “do-not’s” in the process (details to come in my next post).

Step Two: Selection

The second step of the framework is to apply the 80/20 rule by asking:

“What 20% of micro-skills will result in 80% of the outcome I’m trying to create?”

The tagline here is “Material beats Method”. That is, carefully choosing WHAT you learn is more important than HOW you learn that material. Thus, to apply the framework effectively, you must identify and focus on the highest frequency material.

For example, Tim notes that, of the 171,476 English words in the Oxford Dictionary, the 100 most commonly written words (a mere 0.06%) make up more than 50% of all written material.

In the case of Data Science, this has required me to continuously separate the “hot-topics” of today (e.g. deep learning, neural networks) from the core fundamentals (e.g. data cleaning, data wrangling).

Over and over again, I found the following 8 micro-skills (4 technical, 4 non-technical) to be responsible for more than 80% of experts’ results:

The micro-skills I’ll be focusing in on for the next 6 months. More details to come next post.

Step Three: Sequencing

The third step is to lay out the selected LEGO blocks into the most logic progression possible.

Two things to consider here are any dependencies that may exist between the skills, as well as which ones will provide the most early-wins (because humans quit if they’re not having fun).

For instance, Tim grew up 5 minutes from the beach in Long Island, NY, but didn’t learn to swim until he was 31. But what finally did the trick was a program called Total Immersion, with a progression that wouldn’t allow him to fail.

Each exercise was built upon the previous, and failure points like kickboards were completely avoided. Skills were layered on one at a time, and within 10 days he’d gone from a two-pool-length (40 yards) maximum to swimming more than 40 lengths per workout.

With Data Science, I’ve found that learning the basics of a programming language like Python or R is a dependency to the 8 LEGO-blocks listed above. And within each micro-skill, there’s a somewhat obvious progression to learning.

For example, it doesn’t really matter if you start by learning Machine Learning models with clean data or Data Wrangling with messy data. Neither is directly dependent on the other.

However, when learning how to implement Machine Learning models, it’s best to start with simpler algorithms like linear regression and decision-trees before moving on to more complicated approaches like random forests. That is, basic principles tend to carry up to higher-level techniques.

Climbing 2500 feet up Mt. Santubong via dangling ladders & scrambling rocks

Step Four: Stakes

The fourth and final step is to build in consequences and rewards for yourself that will ensure you actually do what you say you’re going to do.

As Tim explains, “if you were to sum up the last 50 years of behavioral psychology in two words, they would be: LOGIC FAILS.” No matter how good a plan is, or how sincere our intentions, humans are horrible at self-discipline.

This is something that I’ve struggled with immensely in the past. As an insatiably curious person, I’ve often found myself jumping from one project to the next as motivations change day-by-day.

So I’ve decided that, if I can stay focused enough to land my first PAID Data Science freelance/contracting/consulting gig by Dec 1, I’ll reward myself with another solo backpacking trip over Winter Break. This one to Patagonia.

Why I’m Doing This

My interest in mastering Data Science is entirely driven by two motivating factors:

  1. Social Impact

One year ago, I published a personal narrative about my recent road-to-recovery from depression. It was a non-traditional path that began through daily mindfulness meditation practice with the Calm app.

Meditating with Calm

Since then, the scientific efficacy in the non-clinical mental health space has only deepened. Day by day, it’s becoming increasingly clear: taking a pill with side-effects or waiting in line for therapy is not always necessary. In many cases, building healthy habits (e.g. meditation practice, physical exercise, diet-change) is a better long-term solution to mental illnesses like depression and anxiety.

This summer I took my first step in this direction by going to work for Calm.com, Inc. in San Francisco. Alongside one of my best friends, we developed & launched Calm College: the first US platform geared toward using mindfulness to improve mental health on college campuses throughout the country.

And thus far, Calm College has launched at 8 schools this Fall — Harvard, Princeton, NYU, Northwestern, Cornell, Johns Hopkins, USC, and the University of Pennsylvania.

Calm College’s Harvard University landing page

Now back on campus, I have two semi-fleshed-out ideas for how Data Science could be applied to mental health and mindfulness:

  1. Use national survey data (e.g. Healthy Minds for college student mental health) to build a predictive model that identifies high-risk students who need a helping hand.
  2. Survey existing users of particular interventions (e.g. Calm) to identify which demographics benefit most from mindfulness-based interventions and match people to solutions accordingly.

I’m still fleshing out these two ideas (and considering many others) so I’d love to hear what you may have in mind?

2. Freedom & Growth

My second motivator comes from this simple truth: my three backpacking trips abroad have taught me more than anything else I’ve ever done.

Even just my most recent trip to Southeast Asia has already caused countless lifestyle changes back home. In two such examples, I’ve been regulating my use of technology by putting my phone on Airplane Mode for 12 hours/day, and just last week I donated half my clothes/possessions to GoodWill.

In essence, I’m loving all the ways travel has helped me grow and I don’t want it to stop anytime soon. So by learning Data Science (the most high-demand craft of the 21st century), I’m earning my freedom to work/live wherever I want after graduation.

Temple-hopping (the new bar-hopping) in Siem Reap, Cambodia

Moreover, I hope to face my fears while building skills that add real-world value. In the best case, I’ll achieve total financial independence. And in the worst case, I’ll have learned the art and discipline of independent/accelerated learning — a skill that’s transferable to anything I hope to learn in the future.

But it’s not just about my own growth. I want to bring you along for the ride. By blogging about my experiences over the next 6 months, I hope to empower anyone else who’s interested to learn along with me.

That’s why I’ll be open-sourcing every single resource, insight, and finding I come across over the next 6 months, through this blog. I hope to build off Tim Ferriss’ framework by creating my own: a free framework for anyone — now or in the future — to master Data Science in less than 6 months.

Already, learning as much as I have since returning from Southeast Asia has made me feel like Superman. And that’s exactly how I want you to feel too: a powerful being in charge of your own destiny.

The Road Ahead

My goal is straightforward: by the time that I graduate college in 6 months, I aim to be a world-class (top 5%) Data Scientist; as measured by the caliber of my professional project portfolio.

Doing this is as much about pursuing my purpose of helping people live more mindfully as it is about empowering others to learn and grow along with me.

So if you’re interested in following along this journey, please feel free to drop your email in my blog’s sign-up bar: https://ajgoldstein.com/

By signing-up, you’ll receive one (just one) email every couple weeks when I’ve posted a new update. And of course you can opt-out anytime.

In my next post, I’ll be going into more detail around the LEGO blocks of Data Science deconstructed, actionable tips & tricks from interviews with experts, a bootcamp I’ve already enrolled in, conferences I’ll be attending, and much more.

Let the fun begin :)

Follow along the journey via the original blog posting:

If you enjoyed this story, please click the 👏 button and share to help others find it! Feel free to leave a comment below.

The Mission publishes stories, videos, and podcasts that make smart people smarter. You can subscribe to get them here. By subscribing and sharing, you will be entered to win three (super awesome) prizes!



AJ Goldstein

Data Scientist. Podcast Host. World Traveler. Part-Time Philosopher.