Nicole Ohanian

My Thoughts

  • // What I did differently

    Daily writing prompt
    What could you do differently?
    View all responses

    I find this to be an interesting question, because I do a lot of thinking about what I “should” do. However, thinking about what should be done doesn’t mean the potential changes you’d need to implement are always actionable. I’ll take a moment to dive into a recent example of a decision to use AI to implement a change in my regular planning/decisioning process.

    When planning and doing things, I tend to get stuck in decision-making for “non-essential” long-term tasks. I have a hard time breaking down tasks or prioritizing in a way that feels like I can immediately start learning and doing. As a result, this often leads to a specific type of paralysis and procrastination.

    I have a lot of goals for this year – The Year of 2026. It’s different now, though. I feel it this time in a way I haven’t for a while. I feel ready to create, to learn. I feel focused and driven and a desire to invest in myself.

    I made a list, this time writing down everything I want to do in the next year and marking the priorities. Yet, I knew that wasn’t granular enough, and a good friend of mine confirmed it. I had to choose what of these things I wanted to accomplish in a much smaller frame of time. She recommended I use Claude AI to help me break down my tasks into a granular 3-month timeline, based on a book she had read, The Twelve Week Year.

    Initially, I had hesitation. It’s not that I never use AI, but I try not to. I really do enjoy challenging myself and doing things on my own when I can. However, how much were my principles getting in the way of getting anything done at all? For example, in an earlier iteration of this thought process, which you can see in an earlier post, I used this thinking to develop a curriculum, but I couldn’t get myself to sit down and really do it. I had outsourced the resource-finding, but I hadn’t done the internal work to figure out how necessary it was for me to go into Data Science to that depth at the time in my life. Additionally, I still wasn’t able to break it down into a schedule that guided me in a way that felt “do-able.”

    So here I am, trying again and aiming to use the tools at my disposal to help myself, preventing arbitrary principles from hindering me. If my weak point is the aforementioned task prioritization and breakdown, why not work with myself instead of against myself and accept the help?

    So, I had a short conversation with Claude to break down my 3 months with my work and my personal life in mind. I asked it to block out my days and weeks in a way that was realistic to me and my tendencies, and? I feel excited. I know I can do this.

    In short, if I did anything differently, it was to work with myself instead of against myself. It was to acknowledge my shortcomings and ask for help and to use the tools at my disposal to make things a little easier when things felt impossible.

    Much love. ❤

  • // A lesson I wish I had learned earlier in life…

    Daily writing prompt
    Share a lesson you wish you had learned earlier in life.
    View all responses

    I think a lesson I wish I had learned earlier on in life is one that I’m still actively learning now – to take myself less seriously in some ways. I’m a pretty fun-loving person. I love to laugh! However, whenever it comes to my creative or intellectual endeavors, I find that I am often paralyzed by a fear that ironically keeps me from starting my practice too often – the fear of not being good enough, and, on top of that, never getting to be good enough. To refer to a book I’ve read that helped me flesh out this feeling, 4000 Weeks: Time Management for Mortals, I believe this mostly has to do with my fear of finitude. How, you may ask, does one have to do with the other? The answer is in my indecision. I procrastinate on making a decision about how to spend my time, further delaying the closure of the abundance of paths ahead of me as I keep my options open. This, of course, is only an illusion. Not making a decision is a decision in itself, and a limiting one at that. One must forge path without the worry that they’re making the wrong choice, because even in avoiding it, they’re probably making the wrong choice to begin with.

  • // The beginning…

    Well, I’ve finally decided to embark on my Data Science journey. At first, I was considering a Master of Science in Data Science and/or Machine Learning-related fields. I looked into a number of programs, whether they be online or through a state university. The goal was to enhance my knowledge in a structured environment without ending up in piles of debt. Afterward, I considered the certification route, looking at accredited institutions as well as the plethora of online resources one has at their disposable thanks to the internet.

    It was a bit of meandering path to make it to the decision I did, but I eventually decided that it’d be best to see how I like a fast-paced, self-driven online course, such as DataCamp. I read many positive things about the resource itself, many swearing by it with their newfound success after gaining the skills it promises to teach you. Given that it had a pretty affordable deal of about $165 as an annual fee, I figured, what was there to lose? Sure, it’s not an accredited institution, but I already have a degree from a great university as well as years of work experience in tech. In fact, I made a mid-career pivot into technical work with a full-stack development bootcamp called App Academy. I’m proud of what I’ve accomplished, and, rather than throwing myself into another pressure cooker, I figure, why not test out my interests in a relatively lower stakes environment with consideration towards my other interests and life obligations. I figure a little less stress couldn’t hurt anyone. This is supposed to be a fun foray into a tangential technical subject that’s piqued my interest. I have this inkling that I’ll really enjoy it if I give myself the chance to.

    Have you gone into data science yourself? If so, what route did you take to start with?

  • // It’s been a while

    Since I’ve written, and frankly? I have still not began embarking on my data science journey. I would say life got in the way, but I can’t really say that metaphor speaks much to me anymore. I believe most things in life ebb and flow, including your interests & priorities. Sometimes, you have to stop trying to control every aspect of your life and telling yourself about all the things you “should be doing.” Maybe some introspection will lead you to realize what’s really important in that moment of your existence, and you’ll focus where it matters to you.

    That being said, I am feeling more focused on my creative and intellectual pursuits than I have for a long time. I’ve learned that it’s okay to settle down in some aspects of your life so you can grow in others, and that’s what I’m going to do now that I feel a bit of relaxation.

    Looking forward to it.

  • // Data Science Curriculum

    So, I gave ChatGPT the following prompt in order to get myself started in thinking about how I would like to pursue learning about the field of data science:

    Hi CGPT, I need help developing a curriculum to learn Data Science. Now, before you go ahead and give me the whole shebang, I want you to know a few things about me and my experience. I’ve been in the tech industry for about 4 years, working in front end development to customer facing technical roles. I have knowledge in programming fundamentals, and don’t wish to review them in a comprehensive manner. However, I do want to understand the foundation of data science so I can decide which branch I want to pursue, and if I want to pursue this field at all. So, for the curriculum, can you consider that and also add helpful resources?

    In case you’re wondering why I did this instead of researching the resources myself, it’s because I want to take the decision-making out of the equation. That is to say that I can get stuck in the planning phase of an endeavor, and decision paralysis will wear my will to start down before I’ve even begun. There’s just too much out there sometimes, and it helps for it to be filtered down and branch out from it later with additional resources. That’s a Pro-Tip if you’re like me at all.

    Now, to be fair, my prompt is a bit specific to me, but you might find that it is beneficial to you if you already have some experience in programming and tech, and wish to pursue DS like I am now.

    This is the curriculum that chatGPT spit out for me:

    1. Introduction to Data Science

    Objective: Understand what data science is, its applications, and its various branches.

    • Overview of Data Science
      • What is Data Science?: Understand the process of extracting insights from data through statistical analysis, machine learning, and data visualization.
      • Applications of Data Science: Industries like healthcare, finance, marketing, etc.
      • Key Roles in Data Science: Data Analyst, Data Scientist, Machine Learning Engineer, etc.

    Resources:

    2. Key Concepts in Data Science

    Objective: Learn the fundamental concepts that form the basis of data science.

    • Data Exploration and Cleaning
      • Understanding the importance of data preprocessing (cleaning, transforming, handling missing data, etc.)
    • Statistical Concepts for Data Science
      • Basics of statistics: Mean, Median, Mode, Variance, Standard Deviation, and Probability theory.
      • Distributions (normal, binomial, Poisson, etc.)
      • Hypothesis testing and p-values.
    • Data Visualization
      • Principles of good data visualization.
      • Key charts: Histograms, scatter plots, box plots, bar charts, and heatmaps.

    Resources:

    3. Programming for Data Science (Python or R)

    Objective: You’re likely already proficient in programming, so the goal here is to get you comfortable with tools used for data science.

    • Python for Data Science
      • Key libraries: Pandas (for data manipulation), NumPy (for numerical operations), Matplotlib/Seaborn (for data visualization), Scikit-learn (for machine learning), and Jupyter Notebooks.
    • R for Data Science
      • Basics of R, dplyr (for data manipulation), ggplot2 (for visualization), and caret (for machine learning).
      Given your background in programming, Python may be more intuitive.

    Resources:

    4. Data Science Workflows & Projects

    Objective: Learn how data science projects unfold and gain hands-on experience.

    • Data Science Workflow
      • Problem definition, data collection, data cleaning, data exploration, model selection, evaluation, and communication of results.
    • Project Development
      • Work on a small project, such as predicting house prices, classifying images, or analyzing a dataset of interest to you (health data, tech data, etc.).

    Resources:

    • Kaggle: Participate in beginner-friendly Kaggle competitions and work through datasets (e.g., Titanic dataset, house price prediction).
    • Project Idea: Use a dataset from UCI Machine Learning Repository and apply your skills.

    5. Machine Learning Basics

    Objective: Gain an understanding of how machine learning works and the common algorithms.

    • Supervised Learning
      • Linear Regression, Logistic Regression, Decision Trees, Random Forests, Support Vector Machines (SVM).
    • Unsupervised Learning
      • K-Means Clustering, Principal Component Analysis (PCA), Hierarchical Clustering.
    • Model Evaluation
      • Cross-validation, bias-variance tradeoff, overfitting vs underfitting, confusion matrix, precision, recall, F1 score.

    Resources:

    6. Exploring Specialized Branches of Data Science

    Once you’ve built a solid foundation, you can explore different subfields to determine which one excites you the most. Here are some branches to consider:

    • Deep Learning
      • Neural networks, CNNs, RNNs, and frameworks like TensorFlow and PyTorch.
    • Natural Language Processing (NLP)
      • Text processing, sentiment analysis, topic modeling, and language models (BERT, GPT).
    • Data Engineering
      • Focus on building data pipelines, working with databases, big data tools like Hadoop, and cloud platforms (AWS, GCP).
    • AI and Reinforcement Learning
      • Explore autonomous systems, decision-making models, and game theory.

    Resources:

    7. Soft Skills for Data Science

    Objective: Build the soft skills necessary for data scientists to communicate insights effectively.

    • Storytelling with Data
      • Communicating results clearly through data visualization and reports.
    • Collaboration
      • Working with stakeholders, understanding business problems, and contributing to cross-functional teams.

    Resources:

    8. Final Project and Portfolio

    • Build a data science portfolio showcasing your skills through projects (e.g., a predictive model, data analysis report, or visualization dashboard).
    • Participate in Kaggle competitions and share your notebooks on GitHub to demonstrate your practical experience.

    Resources:

    I am using this breakdown as a template for how to pursue my studies. As stated in my previous post, for the introductory portion of this I am reading “Data Science from Scratch,” by Joel Grus, which will touch on a number of the sub sections above and provide me with a solid foundation to look into each subject more deeply and give me the vocabulary to understand where I have gaps in knowledge.

    My intention is to read as much as I can each week and do an analysis of what I learned, what I need to learn, and what additional resources I looked into.

    I do tend to take copious amounts of notes when I learn, but my goal with documenting my voyage so to speak is to provide insight into my experience as well.

    Let me know if you are considering doing the same thing in the comments, and certainly let me know about your path as well!

  • // Hello, World

    I tried to stop myself in many ways from setting out to do the things that I wanted to try in life, piling on strict requirements and obstacles in my path. With every step, perfectionism was slowing me down.

    Yet, try as I might, I found myself left with an ember that could not be put out, begging to be fed. Even throughout the coldest of winters, there is a will to survive, and much more than that, to grow. And so I do! I take this as the first of many steps to pursue my dream to write creative and educational content as means to keep myself accountable, but also to help others learn.

    Feel free to join me on my journey of learning about Data Science. I am currently working through the book, “Data Science from Scratch,” an O’Reilly book. I aim to document my path as I continue to learn about different aspects of technology and work on projects.