Close Menu
    Trending
    • Enabling small language models to solve complex reasoning tasks | MIT News
    • New method enables small language models to solve complex reasoning tasks | MIT News
    • New MIT program to train military leaders for the AI age | MIT News
    • The Machine Learning “Advent Calendar” Day 12: Logistic Regression in Excel
    • Decentralized Computation: The Hidden Principle Behind Deep Learning
    • AI Blamed for Job Cuts and There’s Bigger Disruption Ahead
    • New Research Reveals Parents Feel Unprepared to Help Kids with AI
    • Pope Warns of AI’s Impact on Society and Human Dignity
    ProfitlyAI
    • Home
    • Latest News
    • AI Technology
    • Latest AI Innovations
    • AI Tools & Technologies
    • Artificial Intelligence
    ProfitlyAI
    Home » Don’t Build an ML Portfolio Without These Projects
    Artificial Intelligence

    Don’t Build an ML Portfolio Without These Projects

    ProfitlyAIBy ProfitlyAIDecember 10, 2025No Comments8 Mins Read
    Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
    Share
    Facebook Twitter LinkedIn Pinterest Email


    A is the brand new resume — it’s what substitutes for actual work expertise.

    However proper now, your tasks are both ineffective filler otherwise you’re merely not taking them critically, and that’s why you’re not touchdown interviews.

    So on this , I’ll break down the important challenge varieties that top-tier firms truly search for, so you possibly can cease submitting dead-end functions and begin scheduling interviews.

    Let’s make your portfolio the interview magnet it must be.

    3-5 Easy Initiatives

    Absolutely the baseline on your portfolio is 3–5 “easy” or “simple” tasks.

    This is not going to essentially transfer the needle in getting employed, however it is going to give your portfolio preliminary weight.

    Consider these easy tasks because the “warm-up reps” on the health club. They aren’t the heavy lifting that builds severe muscle, however they set up the elemental mechanics, consistency, and self-discipline wanted earlier than you deal with the principle problem.

    The first purpose of those tasks is to get you creating and constructing with no guided tutorial, and to actually get you considering creatively about find out how to clear up issues.

    It’s additionally about “optics” and guaranteeing that your resume, GitHub, and LinkedIn profiles seem lively and well-populated.

    Nevertheless, do take a few month to construct these smaller tasks, guaranteeing they’re of ample high quality and never rapidly generated with ChatGPT.

    Goal to construct a variety of tasks, every utilizing completely different instruments, datasets, and machine studying algorithms.

    If you’d like some inspiration, take a look at this repo I made almost 5 years in the past, which accommodates examples of those easy tasks after I was making an attempt to get my first job.

    GitHub – egorhowell/Data-Science-Projects: A selection of small Data Science Projects.
    A selection of small Data Science Projects. Contribute to egorhowell/Data-Science-Projects development by creating an…github.com

    One thing I will say is that these projects are probably below par by today’s standards, as the field is becoming increasingly competitive.

    So, below is a list of key objectives that your simple projects should meet to make them worthwhile:

    • Variety of Algorithms — try to include Gradient Boosted Trees, Neural Networks, and clustering algorithms like K-Means and DBSCAN in your tasks.
    • Novel Knowledge — It’s a lot better to acquire a messier and extra reasonable dataset that displays the information you’ll encounter in the true world. This may impress employers and interviewers much more, immediately demonstrating your knowledge science and machine studying expertise.
    • Private — To determine what your tasks needs to be on, it’s greatest to begin by answering particular questions you assume shall be attention-grabbing to find from the information. A private contact is all the time higher.

    Finish-To-Finish Mission

    If you wish to work in machine studying, you want to have the ability to deploy your algorithm.

    “A mannequin in a Jupyter pocket book has zero enterprise worth”

    You have got in all probability heard this sentence from me and others a number of instances.

    Having probably the most subtle, fanciest state-of-the-art transformer mannequin means completely nothing until it’s making real-life choices.

    Corporations and hiring managers know this, and albeit, all they care about is whether or not your mannequin is saving or making them cash and whether or not their underlying revenue is growing. 

    It’s actually that reductive.

    So, you need to showcase to potential employers that you know the way to construct and ship an algorithm end-to-end in your portfolio.

    Your challenge ought to ideally embody the next:

    • Knowledge assortment and storage.
    • Knowledge preprocessing.
    • Mannequin coaching and analysis.
    • Mannequin deployment (by way of API, internet app, VPS, and so forth).
    • Evaluation and presentation of your outcomes.

    This challenge is usually the toughest for newcomers to create as a result of it does require some up-skilling and studying a little bit of software program engineering.

    A few of the issues you will want to study are:

    What I don’t need you to do is get intimidated and overwhelmed by the listing.

    Begin small and study the necessities as you go; you actually is not going to want to make use of every thing I simply talked about.

    And as all the time, make it as private as doable; this can maintain you motivated, and it’s a a lot better speaking level in interviews.

    If you’d like a real-life instance, then take a look at one in all my earlier YouTube movies the place I stroll by way of a whole end-to-end challenge I created that forecasts inventory costs after which optimises my portfolio.

    Analysis-Focussed Mission

    I typically advocate that individuals add some analysis factor to their portfolio.

    One technique is to re-implement a analysis paper they’re concerned with.

    You’ll study a lot from this course of:

    • Perceive complicated maths related to cutting-edge fashions.
    • Implement subtle fashions from scratch or utilizing easy libraries.
    • To assume creatively and apply your individual data to new concepts.
    • Enhance your understanding of present traits within the subject and what prime researchers are engaged on.

    And the most effective half is that almost all, actually 99%, of candidates should not doing this, so you’ll immediately stand out.

    Some helpful web sites to seek out papers:

    Re-implementing a paper may be very arduous. I’ve tried a number of instances prior to now, and I nonetheless couldn’t fairly get it 100% appropriate, however I discovered a lot from that course of.

    One other method so as to add analysis into your portfolio is thru studying and distilling papers both by way of writing about it on-line and even by way of a journal membership.

    The latter is what I arrange at my earlier firm, and it was helpful. I offered a wide range of papers akin to:

    It taught me find out how to translate among the most technical subjects on this planet in the intervening time right into a digestible 1-hour presentation.

    This can be a ability that firms actually need, as many practitioners within the subject don’t have it.

    If you happen to at the moment don’t work in an organization the place you possibly can arrange one thing like this, there are numerous Discord and neighborhood teams on the market.

    One group I like to recommend is Yannic Kilcher’s Discord. He’s a machine studying researcher and engineer who creates YouTube movies breaking down analysis papers.

    Write Technical Articles

    Most individuals assume their articles must be “groundbreaking.”

    What if I informed you that’s simply an excuse, and your weblog doesn’t must be distinctive to land you a job?

    If you happen to have a look at mine, many of the posts are about elementary statistical, knowledge science and machine studying ideas.

    Thus far, I’ve written properly over 150 technical and over 60 career-based recommendation articles.

    These began off purely for myself as a way to study extra in regards to the subject; I didn’t care if folks appreciated them or not, as they have been solely for me.

    That is the angle it’s best to have as properly.

    Begin by documenting what you might be at the moment studying or need to study. No must overcomplicate it.

    Having a weblog brings so many positives to your profession and talents:

    • Solidifies understanding of ideas.
    • Helps you assume and have higher communication expertise.
    • Demonstrates a self-starter angle and an curiosity within the subject.
    • Will actually land you jobs and interviews. This occurred to me!

    Your weblog is a passive revenue generator on your profession. The sooner you spend money on it, the higher the payoff.

    I like to recommend you begin running a blog right here on In the direction of Knowledge Science, because it’s very simple to make use of, has a big knowledge science neighborhood, and already has an in-built viewers.

    There are different, extra developer-focused platforms, akin to Hashnode, or you possibly can even weblog by yourself web site, utilising platforms akin to WordPress or Ghost.

    You possibly can even have your individual weblog that you simply create from scratch utilizing HTML, CSS and JavaScript!

    If you wish to study extra, I’ve a complete put up about find out how to begin and write a technical weblog which you could take a look at beneath:


    Now that the precise tasks that flip your portfolio into an interview magnet, there’s only one ultimate piece of the puzzle: the way you current it.

    Most individuals simply throw a GitHub hyperlink on their resume and hope for the most effective, however in the event you try this, you’re lacking out on an enormous alternative to spotlight the enterprise worth of your work.

    To study precisely find out how to showcase your portfolio, see one in all my earlier posts beneath.

    I’ll see you there!

    One other Factor!

    Be part of my free publication the place I share weekly ideas, insights, and recommendation from my expertise as a practising knowledge scientist and machine studying engineer. Plus, as a subscriber, you’ll get my FREE Resume Template!

    Dishing The Data
    main pagenewsletter.egorhowell.com

    Connect With Me



    Source link

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleKey Execs Depart from Apple as It Scrambles to Catch Up with AI
    Next Article Anthropic’s IPO Plan and the Interviewer Tool Set to Change Research
    ProfitlyAI
    • Website

    Related Posts

    Artificial Intelligence

    Enabling small language models to solve complex reasoning tasks | MIT News

    December 12, 2025
    Artificial Intelligence

    New method enables small language models to solve complex reasoning tasks | MIT News

    December 12, 2025
    Artificial Intelligence

    New MIT program to train military leaders for the AI age | MIT News

    December 12, 2025
    Add A Comment
    Leave A Reply Cancel Reply

    Top Posts

    Implementing the Hangman Game in Python

    August 28, 2025

    Building a Сustom MCP Chatbot | Towards Data Science

    July 10, 2025

    Is Your Training Data Representative? A Guide to Checking with PSI in Python

    September 10, 2025

    What the Most Detailed Peer-Reviewed Study on AI in the Classroom Taught Us

    May 20, 2025

    How Microsoft Power BI Elevated My Data Analysis and Visualization Workflow

    May 27, 2025
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    Most Popular

    AI Hype: Don’t Overestimate the Impact of AI

    November 11, 2025

    Multiple Linear Regression Analysis | Towards Data Science

    May 23, 2025

    Data Annotation Techniques For The Most Common AI Use Cases In Healthcare

    May 13, 2025
    Our Picks

    Enabling small language models to solve complex reasoning tasks | MIT News

    December 12, 2025

    New method enables small language models to solve complex reasoning tasks | MIT News

    December 12, 2025

    New MIT program to train military leaders for the AI age | MIT News

    December 12, 2025
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About us
    • Contact us
    Copyright © 2025 ProfitlyAI All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.