While GPT-3 has been released quite some time ago with much buzz due to it’s remarkable capability in writing human-like stories and poems, it has never occurred to me that the API that came with it would provide such flexibility and ease to build data products with wide ranging applications.
In this article, I try to explore some of the use cases that relates to problems that I’ve seen in the job marketplaces — and try to understand how building language based data products might just revolve around ‘prompt engineering’ in the future.
In this article, I summarize my learnings on the idea of growth hypothesis, a stage where startups/companies discover the best channel to grow its customer base after already validating it’s product-market-fit. They’re mainly based on the lessons from GoPractice, a role-based simulator that covers on how to build and grow products.
It’s essentially the assumptions that we make to find the best channel for us to distribute our products. Keep in mind that at this point we already have validated that people have a need for our product, and now the onus is on us to spread the word.
For those that frequent my blog, you’d know that I mostly write about data science, machine learning, and sometimes — the occasional book review. This post isn’t going to be about that (unfortunately?).
This post will talk about how to organize a Usability Test, essentially defined as assessing whether our product’s target audience are capable of completing certain tasks within a prototype/mock environment that we’ve developed.
For context — I’m not a trained UX person, nor have I any experience in this subject matter. The opportunity to work on this came about after my recent job change, where the folks…
This article serves as a summary of this tweet (below) by Sam Bowman. Much of the content is taken from the responses from the original thread, interlaced with my own experience.
If you’re familiar with building machine learning models, either at work or as a hobby; you’ve probably come across the situation where you’ve built tons of different models, having different code bases, and tons of graphs or notebooks and metrics that you need to keep track of as you optimize your code and tune your model to move up that accuracy ladder.
You’re not alone. 
I think over the years as I’ve practiced data science I’ve came across a lot of ways that people tend to deal with this problem. My go to method would usually be an excel…
In my earlier post (Understanding Entity Embeddings and It’s Application) , I’ve talked about solving a forecasting problem using entity embeddings — basically using tabular data that have been represented as vectors and using them as input to a neural network based model to solve a forecasting problem.
This time around though, I’ll be doing the same via a different technique called Random Forest.
Since I don’t intend for this post to be a tutorial on Random Forest, any interested readers keen in diving into the matter can learn more about it in the excellent machine learning book currently being…
The following are my personal notes based on the Deep NLP course by Oxford University held in 2017. The material is available at .
Word Vectors : Representation of word in vector format.
Lexical Semantics : Analysis of word meanings and relationship between them.
Neural network requires vector representation as inputs. Thus there is a need to change words or sentences into the vectors.
Text are merely sequences of discrete symbols (i.e words). A simple way to represent them is by one-hot encoding every words in the sentence. …
As of late I’ve been reading a lot on entity embeddings after being tasked to work on a forecasting problem.
The task at hand was to predict the salary of a given job title, given the historical job ads data that we have in our data warehouse. …
Ever since diving into Natural Language Processing (NLP), I’ve always wanted to write something rather introductory about it at a high level, to provide some structure in my understanding, and to give another perspective of the area — in contrast to the popularity of doing NLP using Deep Learning.
Given a sentence, traditionally the following are the different stages on how a sentence would be analyzed to gain deeper insights.
At this stage we care about the words that make up the sentence, how they are formed, and how do they change depending on their context. …
So I bought this book quite awhile back during a Big Bad Wolf event late last year, thinking that it has something to do about killing unicorn in the unicorn == Uber sense. As it turns out, the unicorn that the author had in mind was essentially ideas that are so fantastic and out of this world that it can’t ever be implemented in real life — which basically wastes everyone’s time and the company’s resources.
In my previous book review, I kinda went and wrote quite a rather lengthy review on what I thought about a book, and how…