My Honest Advice to Beginner ML Students for 2025

[@borismeinardus]

Want to ask specific questions and talk about your ML journey? Let's chat! 👉🏻 calendly.com/boris-meinardus/consulting

[@DataGeek_JJG]

Just started my masters in AI and data engineering at a European university after spending the past year watching your videos from Africa. Safe to say I’ll be rewatching a lot of them for key takeaways, keep inspiring!!

[@borismeinardus]

That sounds so cool! I hope you enjoy the time studying and push through the hard parts haha Imo, it'll be worth it :) Have fun and keep it up!

[@extremexplorer8930]

Borris You were Such a Great Help in my ML journey

[@borismeinardus]

I am really really happy to hear that 🥰

[@Channu-c7e]

3:40 now I'am in that phase. Thank you Boris for boosting my confidence🙂

[@borismeinardus]

I am really happy to hear that! That was one of the purposes of this video ☺️ Keep it up and have fun ☺️

[@EduardoVodopives]

All of Boris's advice is great. Since I started applying his suggestions from the implementation paper, my practical ML skills have improved significantly. Thank you!

[@borismeinardus]

Wow, super happy to hear that! I'll do my best to create more content that can help :) Keep it up and have fun learning! 😊

[@Gabriel-Aggrey]

your video editing skill is top notch

[@borismeinardus]

Thank you! Nowadays, I have an amazing editor, so all credit to him on that part 😊❤️

[@ankitchoudhary9302]

hey today i just reallsed the same thing and even started working with the project and feeling little confident and your video just validated my thoughts, thanks dude.

[@borismeinardus]

That's awesome! Really happy to hear that! 😊 Definitely keep it up and try to have as much fun as possible! 😊😊

[@mdaniels6311]

I really appreciate what you do brother.

[@borismeinardus]

And I really appreciate all your support ❤️

[@Chadpritai]

Congratulations on getting into sakana ai, share your interview experience and behind the scenes preparation that you helped you land this job 🎉

[@borismeinardus]

Thank you!!! I will definitely soon do an update video where I will share a bit of the behind the scenes. I can't share many details on the exact interview, of course, haha but I'll do my best to share as much as I can :)

[@ShivamKumarSharmaSGEC]

i am right now building an end to end data science project and honestly the advice you gave about project based learning is absolutely true. The more i learn, i realize that there's still a lot more. But i will keep moving forward.

[@borismeinardus]

Great to hear! I also feel like I know nothing haha Just keep going and enjoy the process of learning :) Have fun! ❤️

[@Ari-pq4db]

That's some awesome advice, will definitely look into it. Also could you make videos of making ML projects (Like all the planning, coding all that stuff). I'm planning to start my ML journey from coming December ❄🔥

[@borismeinardus]

Sounds amazing!! I have 2 videos on how I work on projects, maybe those will be helpful! kzbin.info/www/bejne/oYSQl4eDn5yLeqM kzbin.info/www/bejne/d6qqq5l9htZ9nbs Definitely keep it up and enjoy the journey 😊

[@tajinder2317]

Thankyou brother seeking some knowledge just like this 🍀

[@borismeinardus]

I appreciate it!! I am really glad to hear that my simple videos provide at lest some value 😊

[@tajinder2317]

@borismeinardus keep it up bro 🕊️

[@darshantawte7435]

Please make video for students who will be going for masters for MS in AI/ML. What to focus on , how to make most out of it etc.

[@borismeinardus]

I'll hopefully soon make a video where I should cover most of your questions :)

[@Umer_Aftab]

Congratulations bro. U are an inspiration

[@borismeinardus]

Thank you so much! I just try to share my journey and what I learn along the way and am super happy that people gain value from that ❤️

[@Umer_Aftab]

@@borismeinardus You're doing an amazing job! Your journey and insights inspire me and so many others. It's refreshing to see someone genuinely sharing their experiences and knowledge. Keep shining!

[@BrianDevJourney]

Thanks for spreading the knowledge to the students of the future. Emailed you about a Sponsorship opportunity. Cheers!

[@Sauvage-xk7he]

Congratulations bro for getting placed as Research Scientist at Sakana ai 🎉

[@borismeinardus]

Thanks a ton! I will hopefully soon make a video on that topic :) A "Life update" style video haha I had some hectic few weeks haha

[@MasaoTaketani]

@@borismeinardus Oh, really? You're gonna work for Sakana AI? Congratulations!!! I'm happy for you!!!🥳 Does that mean you're coming to Tokyo? If so, I'd like to meet and talk to you in person at some point. I've been working as a deep-learning-based AI engineer for a while in Tokyo. So, really nice to know someone who is also super passionate about the field. Welcome to Tokyo, Japan! And good luck with your journey at Sakana AI!!!😆

[@maazahmad3988]

What kind of projects should we build and how to get ideas for them? As for software enginnering, skills, do we need to learn them seperately or they will naturally come during building projects?

[@borismeinardus]

That's a difficult question. Generally, you would get ideas by looking at what others do, try to recreate that and by doing so, get and understanding for what is possible with the technology and what it's current limitations are. In general, the more you do, the more you learn. So the best way to find a good project, is to just play around with things you know of and look at what others are doing. That can be done by looking at open source software and reading papers/ blog posts. Kaggle might also be a good starting point. Another point to get inspiration, could be this link: cs230.stanford.edu/past-projects/ I hope this somewhat helps. Just don't get frustrated and give up! Keep it up!! :)

[@nuobeitang]

Thanks a lot❤But it would be better if you could share some experience as an ml engineer and let us know what an ml engineer do in daily work🎉

[@borismeinardus]

I will hopefully be able to record a "A day in my life" style video :) Please note, that I am a Research Scientist, so it's not the same as an ML Engineer when it comes to the content of my work, but the layout of the day is probably very similar :)

[@daskoujr2670]

you had a video talkinb about sentiment analysis project but i cant find it.

[@dhruvakalva9911]

Where can i find Andew Ng courses in Coursera?

[@borismeinardus]

www.coursera.org/specializations/machine-learning-introduction www.coursera.org/specializations/deep-learning

[@i_youtube_]

Brois, I want to specialize in computer vision and NLP. Should I learn a lot about ML? Any advice?

[@borismeinardus]

Absolutely. The ML basics are necessary for every single one. At least, in my opinion and experience. They are not the most flashy things, but core knowledge, that actually is very elegant if you give it a chance haha

[@Benhuyaaawr]

while doing real-time based projects, I found ML quite different from theory, thoughts?

[@borismeinardus]

not sure what you mean by different from theory, but ML theory and real world ML are indeed quite different w.r.t the skills you need! That is completely normal :)

[@bansarigaloria6149]

Hello Boris, Thanks you for your amazing videos, suggestions and advices on ML carrear. It really helps. I have studied ML algorithms and now want to work on project. My Question is , is it necessary to do project with End to End implementation, that includes MLOps tools and software? Is understanding of MLOps tools and concepts required for ML engineer?

[@borismeinardus]

Thank you! Sounds great!! In general, especially for a beginner, don't worry too much about the exact details of the project. Really. Your first project won't be the best one. And no project covers absolutely every aspect of what an ML engineer might need to do/ know. For some projects, you might not even need to train a model! There are many papers published at the best conferences, where they just try out different things with pretrained LLMs (again, no training at all!). As long as you can get a cool idea, execute on that, and share what you did and how you did it, you are doing great! Just repeat that until the end of time haha I hope this somewhat helps :) Keep it up!!

[@bansarigaloria6149]

@@borismeinardus Thank You🙏

[@Oluwatimilehin_G]

How do you get projects please?

[@borismeinardus]

There are many different ways. For beginners, I recommend looking at Kaggle! If you are more comfortable, you can think of your own ideas, and read papers to get inspiration for own ideas or simply for reimplementing. I hope this somewhat helps :)

[@goodnessnwanebu8662]

Hey, who’s your editor? I’d love to talk to him/her.

[@SukeerthiAdireddi]

hi i want to do ms in ai , do you recommend it

[@ISaidTheMan]

can you specifically give us ml research content?

[@i_youtube_]

Any plan to launch ML courses? We are ready to pay❤

[@ivonluiz5935]

Your channel is going places bro. You talk really weel to the camera.

[@borismeinardus]

I appreciate that! Just trying to be helpful and improve step by step :)

[@brandeltsagueu5455]

thank you for this very informative video :) Where can I find and access research papers published on the subject Machine Learning ?

[@borismeinardus]

I‘m really glad you liked it!! arxiv.org, there you can find all papers otherwise, have a look at papers published at top conferences, like NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, … I hope this helps :)

[@Alec-FRANZFerrer]

Hey Boris, I am a hs student wondering which program to apply to, I was wondering what graduate program did you study and if it is essential for finding a Machine Learning Job. Can you find ML Jobs as a Data Science major or is Comp Sci/ Software eng better?

[@sutofana]

THANK YOUUUUU 🤍🤍🤍🤍

[@borismeinardus]

🚀❤️🚀

[@youssefkallala-vh4bj]

what is the best course/book to study those fundamentals

[@borismeinardus]

hmm, nowadays I am a big fan of the book "Why machines learn" by Anil Ananthaswamy

[@riteshbhadana]

Bro how analysis a research paper pl bro answer

[@TheIvanpartizan]

Hello Boris, nice video, thanks! Is it too late to start learning at 38?

[@borismeinardus]

Absolutely not! You only have to be aware of the time commitment it requires. People at 38 might have different life circumstances than people at 20. If you can accommodate the time to invest into studying, this is a very good time to get into ML ☺️ Good luck and keep it up!!

[@TheIvanpartizan]

@@borismeinardus thanks Boris, I will try to make it in ML :)

[@vaadewoyin]

This is very informative....

[@borismeinardus]

Thanks! 😊

[@jeanetteb.pedersen8164]

I feel confused. There are a lot of new research but also a lot of projects are failing (85%?). That is the point of spending time on new research if it cant solve problems real world anyway? I have been involved in some ml/data science/ automatization projects and more than one is in production and working. Llm models are not going to give me 99% accuracy which is needed if the ml system looks at the data and sorts it without a human in the process but it is all everyone is talking about. I feel like I am the bad engineer because I would rather spend 3 hours making and algorithm that actually did something rather than 2 weeks setting up llm to do the same but with more unpredictability. But using an llm is now more impressive than actual problem solving skills. I feel like a dinasour stuck in the past.

[@carsongutierrez7072]

Alright mate!

[@borismeinardus]

🫡🫡

[@noadsensehere9195]

Good luck for Sakana AI job.

[@borismeinardus]

Thank you very much ❤️

[@D7Mut]

You said math but what exactly I need as a ML in math because math is like an ocean be specific

[@borismeinardus]

I'm just guessing numbers here, but I think for like 80% of ML you simply need an understanding of vector/ matrix operations, and understanding of how to compute derivatives/ gradients, and an understanding of the basic probability terms surrounding Bayes theorem and gaussian distributions. There is always more that you can delve into and for certain problems you will need more math, but for the most part, what most people (especially non researchers) that should be enough. Those fundamentals will then be applied to different models in different ways, but I truly believe, that those fundamentals are mostly enough. I hope this somewhat answers you question :)

[@GJRahul-rr3uk]

Is SQL important in ML?

[@borismeinardus]

Depends a lot on the role. For many data scientists that do more analysis work, yes! Personally, I used it only a little bit, but that was very limited and I simply googled everything I needed :)

[@AHMADRYANPERMANA]

I found this video from medium🎉