How Can You Become a Machine Learning Engineer?
Вставка
- Опубліковано 30 тра 2024
- In this video, some real-world insight into the machine learning engineering role.
If you're interested in the fastest path to a career in machine learning or any other data role my platform called www.logikbot.com is affordable, real-world, comprehensive and used by top tech companies.
It's affordable. It's less than $20/month right now. I created the data analysts path after I realized that entry level machine learning engineering roles are gone.
The data analyst path is simple. You take a SQL course, you take the PowerBI course and then you study the exam simulator for the PowerBI Certification. You schedule and pass exam DL-300. After that, you're ready to begin interviewing for data analyst roles.
If you're in a data analyst role and are ready to learn real-world machine learning, my platform is a complete solution for you. The platform has several courses that no other one does. The first and most important is the data cleansing class. That alone is worth the price of a three month subscription. Secondly there's the master class on imbalanced classification.
Yes, there are other paths but LogikBot is the most affordable, real-world path out there.
Thanks.
Mike
LogikBot - Affordable, Real-World and Comprehensive - www.logikbot.com
There are two core paths on LogikBot. One for the data analyst role and one for the machine learning engineering role.
I've added the data analyst role recently because machine learning has become so difficult to break into. If you're new to IT, I'd highly recommend this path. You can be in a real-world data analsyt role in under a year if you really apply yourself. LogikBot is the most affordable and fastest way to attain this role. The platform even has an exam simulator for preparing for the exam.
For the data analyst role, you take two courses then prepare for and pass exam PL-300. This exam, PL-300, is the Microsoft Certified PowerBI Data Analyst Certification.
If you're in a shortcut role, then there's an entire path laid out for learning machine learning. The platform used by Uber and Microsoft. I worked at both companies. The courses are real-world and comprehensive. This is not the easy way. There is no easy way. This is one of the few platform that's offers a real-world approach to working in AI.
would you recommend getting a data analyst job first after completing sql course on logikbot and then after take the ML course after getting some job experience or would it be okay to study back to back?
@@1bullgod Exactly. I'd recommend the data analyst role for two years while you're learning machine learning.
mike i do get whats the data amalyst role thats what im gonna aim now.. but what does the shortcut role mean? i didnt quite get it
@@thedatajanitor9537How am i gonna learn machine learning tho so does that mean the even from my bachelors at ai/ml won’t help
I just spotted your channel and used 1 hour to watch your videos. As a newbie data scientist, you are absolutely right. This is what I have been practicing for months now. At first I said this is going to be one of those killjoys and hope-dashers. But I listened and I am liberated and ever more hopeful.
At some point, I know I will become a machine learning or data engineer. There is just so much confusion concerning the roles so I chose the ordinary name Data Scientist. But I agree SQL is required in all of them.
You mentioned every step I am struggling to gain experience with as an amateur who is very worried about his future in the job market.
Sourcing and packaging
Analysis and visualization
Preprocessing (cleaning and normalization)
Modeling
Evaluation
Deployment or production.
You mentioned ensemble models. I saw a post where a narrator claimed ensemble models performed better than neural networks.
AI is about prediction and suggestions based on pattern predictions. To the layperson, it looks like intelligence.
Thank you for the enlightenment
I saw this video. As someone who majored Statistics and Machine learning at school , I can assure you this is the most honest take on machine learning engineering ever. I have been on a job hunt for more than a year already.
Thanks. Have you had any luck? Any positive feedback from interviewers? It sucks. The data science scam has really made entry level roles very difficult.
@Kief ua-cam.com/video/L4NS7o72-E8/v-deo.html
@Kief Nope. Here you go. ua-cam.com/video/J7EraDOPJPU/v-deo.html
this is the reddest pill about data science i've ever seen, thank you sir
Thanks. Nice to have you.
I found you Data Janitor, Michael West straight from Quora. Your 30 years of experience in data speaks volumes. I have watched a number of your videos and found them well-curated, resourceful, and bare truth with no embellishments. You have no time for sideshows and theatrics. I am in my 3rd year of Bachelors in Computer Science. Is it advisable to model my career in Big Data by starting in a Data Analyst role and working upwards? According to most of your videos, one can get an entry-level data analyst job straight from a bachelor's program.I will appreciate your advice on this.
Thanks for this one. What you mention just confirms what I have learnt so far. Just for the sake of sharing, I have been a "machine learning engineer" intern at a financial institution (a microfinance more precisely) for 6 months during which I explored the task of building credig scoring model for existing customers. I have read books, papers, blogs which helped me understand the field of machine learning. During this internship, I have also made so many mistakes during the model building phases. I consider this internship as a "failure" since what I built won't surely work in production but I have learnt a tons.
This is a masterpiece. Straight to the point, confident, and most importantly helpful.
Thank you for those kind words and welcome to my channel.
Thanks for the Complete and most detailed analysis i have found on youtube.
You are welcome! Welcome to the channel.
Wow... Thanks for the straight talk. I really enjoy the frankness. I will will be watching more of your vids.
Thank you Steven.
Hey Mike, you saved me two years of study with all your videos. Thank you so so much. Will sign up for your course
Thank you.
Thank you for making such straightforward content.
Thanks and welcome to the channel.
I watch this video like 10 times
There’s a lot of insight
Thank you. I appreciate that!! I have lots more. :)
TRUTH! Great info here. I will have to check out your platform.
Thank you Lamarr168. Nice to have you!!!
Also thank you so much for your video and guidance.
Love your videos, somebody who is really telling the truth and I love the phrase "the real world".
Thank you!!
Thanks a lot for bashing me with this fact, it makes me more motivated to learn
I really appreciate your video. It did reassure me that what I've been doing is a right bath. Also, I've noted some great information
Thank you!!
This is one of the best advices I´ve found on the internet, I´m pursuing a PhD in Intelligent Systems but didn't get much real world insight from professors, most of them are in Research rather than production environments.
The most important real world insight was when I worked in the electrical industry and realized there is a necessity of data related roles in the industry, nonetheless didn't found where to look on how to become a Data Engineer (Most of the posts and everything is like courses of people who is not in real world applications), is it better to master python and SQL to become a data engineer rather than other languages or frameworks?
Personally I really like Math and Statistics but realized that most of the tasks in PhD of writing regressors and Perceptron from scratch is not necessary for applications since there are many libraries from model teams with very efficient programming and optimization, the critical step is cleansing data and picking the correct model.
Thanks for your videos, any suggest is welcome.
Thank you David. I appreciate that. Here's a course that may help you with data cleansing. www.logikbot.com/courses/data-cleansing-master-class-in-python
@@thedatajanitor9537 where did you learn content of this course ?
@@vamostimow Books.
Thanks for this! Much appreciated
Glad it was helpful!
Great video. I didn’t realize the data scientist role was actually a ML engineering role. I am still undecided as to where I want to take my career but I recently learned about the research side of things and was wondering which position has a bigger impact on a company, a research scientist/engineer or a ML engineer.
There's no comparison. Researchers have almost zero impact on a companies bottom line. Machine learning engineers have a continual positive impact.
Thanks for the compliment and welcome to the channel.
Hey my brother! So this is my goal on how to become a machine learning engineer! Im a freshman in college but imma follow your approach. Its been tough finding roles in this freshman year.
I’ve been unemployed this whole year but this has been great learning experience to see how tough it is to get a part time while studying and this will grow my tenacity when searching my role in being an ml engineer
You won't get a job in machine learning without a college degree. Focus on that first.
@@thedatajanitor9537 thanks im doing that right now, so just focus on getting experience in data analyst roles, and have that punched in these years?
very good info. please keep posting.
Thanks, will do!
I love this guy's honesty.
Thanks and welcome. Most don't. Too many pansies. :)
@@thedatajanitor9537 No worries mate, i'd really appreciate if you could also present your couse on your website in a career track manner. For exmaple, 1 -> 2 -> 3 , so someone like me who is starting out wants to know which course to pick first. It is always better to learn from someone who is cruelly honest than someone udemy instructor.
Thanks for the video. Great info. Very honest.
I have been a SQL developer, BI developer, ETL developer, performance DBA for over twenty years. Currently working as a junior Azure data engineer. BS appl math, MS data science, proficient in python. ML jobs are extremely difficult to get, even a first interview. I might go the data analyst / PBI route to become an ML engineer.
Thanks for sharing!
What? 20 years of experience? BSc in Math? MS in Data Science? Dude wake up, you are overqualified for a data analyst.
@@arseniikvachan Well, yeah, I have been turned down for for jobs for being “overqualified”. Maybe the best path is to get a masters in statistics.
good insight. nice video.
Thanks!
The only cool eye-opening video I have seen....thank you
Thanks Emmanuel. Thanks for watching.
@@thedatajanitor9537 I have registered on logikbot. However, I am 32 and have just basic programming and mathematical skills..do you think sir, with your experience, I can make it if give in time and dedication?
@@criticalthinking2019 I believe you can make it. Don't give up. Break you learning down into digestible segments.
Sir! your teaching method very imprasive.i like it.and it become a very helpful for me.
Keep watching and thank you.
I am pursuing my Master's in Data-Science and I truly respect you for preaching the facts.
Most of my colleagues and me(of course) believe that a Master's degree in Data-Science implies expertise in Machine Learning and to quote you, 'You are in for lot of heart ache' is putting it nicely.
😅
Yeah... most of you are in for one hell of a wakeup call. I'd tell them to plan on being unemployed or working at Starbucks.
Thank you!
Sure. Thanks for watching.
Little harsh, but true for the most part. However, you forget to mention that with a masters or phd (mostly just phd) you can get an Applied Data scientist position, though this is not a machine learning role like you said and is harder to get. Also, these roles are really for the best of the best, so definitely not an entry level job or an easy path. If someone is looking at the MOOC content just starting out and hoping to get one of these jobs, then yeah your advice is what they need to hear.
You are correct. They just fired an entire team of "applied scientists" at Amazon. We don't need more researchers, we need more doer.
Love this video! I just am finishing up my masters in computer science and I really want to get into ML, thank you for the advice!
Sure. Good luck.
@@thedatajanitor9537 great videos, had to come back to this because you’ve been right about data jobs and academia, I was able to start off as a data analyst and I will then work my way up to MLE after having real world experience! This is coming from someone who spent 6 years in academia.
What skills would you generally consider to be important to have for a junior data engineer roles? Thank you for sharing your thoughts Mike!
Did you see this video? ua-cam.com/video/7IOnBdpT0CY/v-deo.html
@@thedatajanitor9537 I did not; thank you for the link!
Honestly I am surprised how you don't have 1 million subscribers . You really get to the important points
Thanks and welcome to the channel.
@@thedatajanitor9537 Haha I am really learning how real life goes on with your videos unlike these guru vids that I have seen of other people .
@@thedatajanitor9537 Unfortunately I am not able to purchase your LogikBot because on my country the dollar value is kinda of high . Mr Mike can you please upload a video on legit free resources to learn to be able to secure a ML entry position and start writing real life ML projects? If you can't can you tell me about the resources in a nutshell?
No such thing as entry level in machine learning. Here... watch this: ua-cam.com/video/svUWW2f_6LY/v-deo.html
@@thedatajanitor9537 Thanks very much for the eye opening video :D . I also saw your How Can You Become a Machine Learning Engineer? . Boy oh boy I never knew data cleansining played such a big part . I will start to focus more on data analysis for now.
Hey Mike, super informative and to-the-point video, much appreciated. I am currently an Electrical Engineering student at well known STEM school. My school offers an EE degree option with focus in Machine Learning and Data Science, and I’m wondering if this is something that sounds worth my time if I have aspirations to work in the ML field, or if it’s just a smoke and mirrors thing and I’d be better off pursuing something like Computer Engineering or even CS, or if they’re all viable means to an end and none is inherently “better” than the other. Keep up the great content!
We don't value degrees in the real-world. However, strangely enough, you need to have one in something. If I were headed into machine learning and needed an undergrad it would be in statistics.
This was a very realistic take on becoming a MLE, I’m switching careers right now. I’m taking a free prep course for a coding boot camp Codesmith CSX. Would you recommend a more efficient route. Codesmith is quite expensive sitting at about 20k.
I'd recommend you set some realistic expectations. Bootcamps are a joke and the top skill you'll need for any MLE role is SQL. I'd suggest the data analyst path. Without SQL skills, it won't matter how good you are at coding, you won't get a job. SQL is king. Here... watch this,.
ua-cam.com/video/OoDAfidxHVQ/v-deo.html
Awesome channel, followed and subscribed. Recent electrical engineer graduate and then did a data science diploma at bootcamp. The information in this 10 mins clip is priceless. Will save you so much time. Find an entry level data eng/data analyst role, gain skills in SQL and ML pipelines. I have a quick question how long do you think it takes to have a decent level of experience to then apply for more machine learning engineer roles on average ?
A year or two should do it. Thanks for the compliments.
@@thedatajanitor9537 Sounds reasonable, thanks again !
@@digitalnomad2196Hey man, thinking about going down the same road as you did. Any advice?
I feel like Vic Mackey just hit me in the face with a book of hard truths then burnt my face on the stove of reality. Thanks!
Sure. Most need it. Most people are fucking clueless... and that's why there are 300K open jobs for MLE not being filled.
Hi, Mike. Great informative video.
I’m starting from scratch and was looking to get a role as a MLE.
But what I understand from your video, it’s best to start out as a DA… learn ML as you build experience in the work force… then take the ML course, then apply.
My question is, does your website teach everything to become a DA then move onto a MLE? And what is the timeline for something like that?
Sorry for the long question and thank you for the information.
No such thing as any platform that has everything you need. That's an impossibility. Simply too much information. My platform is the best starting point for a job as a DA in the real-world.
@@thedatajanitor9537Thank You
I visit you website thats great👏👏
Glad to hear that
Thank you for going straight to the point and for not beating around the bush. Very clear message. I guess you must be good at...INFORMATION CLEANSING as well 😂😂😂
I'll leave now
Hence... the data janitor. :)
@the data janitor thanks I've been stuck on your videos because for the first time someone who knows speaks the truth, I affirm this because I've been on the search for years, at least since after completing Andrew NG's ML course in 2020. I even had to enrol for a 2nd degree in Computer Science only to discover that DA is the gateway. What do you suggest, would the CS degree really be so much of an advantage?
Nope. Degrees are worthless in AI. Watch this.
ua-cam.com/video/LWH3bLDz6Qo/v-deo.html
The only entry level data job I know of is the DA. If you can do it another way... great... I just haven't seen anyone do it.
This was such a good information, recently i started to do my ms in data science, even though i have 3yrs of experience as a software developer, when i looked for jobs they all require experienced people and i was confused did i choose the right path, or what next should i do to get into the right path
Let's hear from you in a video: How did you learn applied statistics and math for machine learning?
On my own. There isn't that much math and most of it's applied.
agree. Math is needed for 2 reasons. 1] To get insight (who cares)
2] Folks who write ML libraries and optimization algorithms, someone like who is part of the Tensorflow or Pytorch team.
loved the videos, one question , I have audit background and a little analytics for fraud, I practiced sql and r for financial data analytics, im starting a masters in data science and I hope to transition to ml jobs for financial firms after a few years, Im very weary of the steps im taking, I don't know if another bachelors in comp sci with heavy data science credits is better or just a masters in data science would suffice? ps: I have a Bsc in accounting and finance
So, as a summary. Entry roles as data analysis or programming and learn python are the best way to start your way up to a Machine Learning position. Any recomendations for someone starting programming in 2022? Thanks
Programming roles are going to require you know how to program. These aren't easy roles to attain. The easiest data role to attain is the data analyst. The easiest IT role is the help desk role. Either way, getting into IT is the hardest part. Once you're in you can move laterally or more easily learn new skills. Yes, any SQL cert will help you.
Thanks for the nice video!
What do you recommend to someone working as a Data Scientist and wanted to transition into ML Engineer role?
The DS role is dead. Here. ua-cam.com/video/14RPhvDN4uo/v-deo.html
I recommend you learn the roles in the real-world. Here.
ua-cam.com/video/5iPgpad37x4/v-deo.html
Hi mike, thank you for always giving real world advice in real world job,
I have a question, If someone want to do the programming path as an entry level role for machine learning, what would you recommend to get the first job as a junior programmer because i don't know any programming certification and i don't have IT background
Whatever junior level programming role you have make sure it includes heavy SQL. Without data knowledge, your programming skills are worthless at most companies. Data is the king in machine learning.
@@thedatajanitor9537 ok thanks again for the advice mike
Do you think asvance degrees like phd would help in a machine learning career? Companies like anthropic demand phd in a lot of their roles, is it the rule or the exception?
I don't know of any companies that demand PhDs outside of worthless research departments. I don't know any skilled machine learning engineers working in the real-world with a PhD. So, no... I don't recommend and it won't help you in the real-world.
I have some quick questions, please. Would you recommend after getting a computer science undergrad, I go for a data analyst role or instead go into a python programming role? What is better? Do you recommend any certifications to compliment the job. Do you think getting a Master's in Machine Learning, only after getting work 2-3 years of experience, is good to make me more valuable? Any help would be greatly appreciated and thank you so so much.
Start watching my videos. I've have hundreds of them. All these questions are answered in them. Every single one.
Great content; I love it.
I'm in the second year of my Ph.D. in Compute Science, and my research is in Deep Learning. I wish I had found your video before I started my Ph.D. It's late for me to drop out of school, though. I also have 3 yoe of coding Python for my cloud security job. I know I will need a lot of SQL.
Do you have any advice for Ph.D. people (AI research focus) to not waste time and start as a machine learning engineer? Thank you.
Nah. If you really wan to do research then you do need a PhD.
I'm doing biostatistics and interested statistical genetics research (not doing independent research yet, since I'm still taking classes) It seems like there's a lot of overlap for ML engineers/researchers and bioinformatics. Don't you get to develop practical skills using SQL and Python/R for your (or your supervisor's) research or are you stuck teaching? Maybe try to find a lab that's more applied if you're not getting that experience
Hello, I have stumbled upon this video in a quora post, and would like to ask for your advice. I'm a first year high school student, that has immense interest in machine learning. To pursue this interest, I've started Andrew Ng's "Supervised Machine Learning : Regression and Classification" course. I've been enjoying it so far, and understanding little by little how things work is very rewarding. So I'd like to ask, what do you think is the best course of action I can take from here? I'd also like to add that:
1. while I'm taking this course I'm self-studying more math, as my current level of math is probably not sufficient.
2. The reason I took the intiative and started learning originates from me starting to self-study python a few months back. I'm fairly proficient with it, and have learned how to use several libaries, mostly related to web scraping . Some examples are: BeautifulSoup, selenium(made some fun bots with it too) and scrapy. Scrapy also led me to studying some basic SQL.
I appreciate this video immensly, and hope you could reply.
Focus on getting good grades in high school. There are no entry level roles for machine learning engineers and *without a bachelors* you have zero chance of working in any real-world machine learning role.
@@thedatajanitor9537 will definitely not sacrifice my grades, nor am I aspiring to get a job right out of highschool, I am just looking to expand my existing knowledge foundations as much as I can: if not for better skills in the future, than to satisfy my own curiosity.
That said, any advise given here is something I'd like to abide by, so what bachelor's degree specifically would be ideal? A computer science degree?
@@iyar220 It won't matter. A degree in stats is a good idea and any degree where you learn to work with data would also be great. If I were you, I'd continue to learn SQL and Python and gain some cloud skills and certifications.
@@thedatajanitor9537 Thank you!
If someone has a basic understanding of python but not enough to be employable yet, in this economy would it make more sense to focus on studying for a data analyst role or a junior back end engineer?
I believe it's much easier to get a DA role than a junior dev role right now. Again, if you are a junior dev and aren't working with SQL then all the Python in the world won't help you get a job. The real-world in ML is about data.
As you said, data engineering role is the top role in all of IT, should I become a data engineer or a machine learning engineer? If I should go down the path of machine learning, should I become a junior data analyst or a junior data engineer?
That's up to you.
Great video! I'm in my last year of undergrad at a top 3 computer science school, trying to determine whether I want to do a masters. I'm specializing in vision, and most companies I have seem posting Computer Vision Engineer (also sometimes called Deep Learning Engineer or Machine Learning Engineer on a vision-focused team) roles require a masters degree + ~2 years of experience. I've also noticed that in the 3 companies I've done internships at, most engineers in these roles had masters degrees or higher.
So, it seems to me like continuing to masters is the right choice, since companies seem to prefer this qualification for roles in computer vision. What is your take on this @the data janitor?
I don't work in that space. Honestly don't know what computer vision or NLP roles require. I don't work with deep learning models. I do know it's only 20% of real-world machine learning so it's insanely competitive. Around 80% of all appleid machine learning is modeling structured data.
I've worked at several top tech companies and none of the top MLEs I worked with had a masters. Not one. There were a few faker scientists but no one took them seriously.
@@thedatajanitor9537 Thanks for your honesty! I've heard the term "faker scientist" used a lot by your audience. Does this just refer to someone with a higher degree that doesn't have the experience & expertise to back it up?
@@christianfoley7441 Exactly.
Although I am graduating with a masters this summer, I now know that it is not an important factor in becoming an MLE. I know the basics of SQL and have experience in Python and ML libraries. Would this still be considered as "starting from scratch" in the real world?
Yep. You are in for a rough ride. I'm not being mean... just honest, degrees in IT don't really matter.
Watch this: ua-cam.com/video/LWH3bLDz6Qo/v-deo.html
I was searching for data science and machine learning courses on online platforms since last year. Here my journey ends now. I am a mechanical engineer and have 6 years of experience in mechanical design and R&D. I am looking to further transition my career to a data science or machine learning engineer. Mike, please let me know that are those courses on your platform will be enough, or do I have to take some other courses as well? Guide me in this journey. Although I have already learned a decent amount of python from Udemy.
No one is going to hire you without heavy data experience. There are no entry level roles in machine learning or data science. You'll need to start off at a more entry level role like that of a data analyst. This will be painful because your salary will take a hit. However, right now... there are no other ways other than coming from a programming role.
@@thedatajanitor9537 ok, can you guide me then that how could I enter into data science? How I start?
@@akshaypanchal424 I would suggest you focus on the data analysis role. It's a great entry level role and many of the machine learning engineer roles require SQL and a tool you'll need for the data analyst role. This video should help: ua-cam.com/video/KFmXhoNrcyQ/v-deo.html
Good afternoon sir,I'm 21 years old amd I'm looking into making entry into data analyst I came across ypte video recently about Starting with data analyst and then making a transition into data engineering I have to say I loved it !!! Sir so if i get a strong foundation in sql and power bi or tableau is it sufficient to get an entry level role in data analyst ?
That's the only option you have. You'll need certifications. For noob they are critically important.
Hey, I’m months away of finishing high school, I’m 17, I consider myself pretty decent at math and statistics, I’ve been into coding for a few months (mainly Python), and by now I think I’ve decided to be an ML engineer. Thing is, due to personal reasons I’ll be forced to go to college, I’ll have the opportunity to choose between CS, CE, or SE, should I just go and choose SE since I think that it would be the least time consuming for me? (Considering that I’ll learn the vast majority of ML knowledge out of college). Thanks :)
LMAO. No, you won't learn the vast of your knowledge about ML in college. College is all theory. Here you go. The myth of academia in IT.
ua-cam.com/video/-AD3S6TC2nY/v-deo.html
@@thedatajanitor9537 I meant to say that I was going to learn the vast majority of ML knowledge OUTSIDE of college, but thanks anyways, I’ll consider getting your courses once I think I’m ready 🙏
You'll do fine. You're watching my videos. You need a degree in something so it might as well be ML or stats.
Hello Sir, love your videos. I am an CS undergraduate senior and did an SWE internship.
Quick Question.
You mentioned becoming a data analyst and then looking into being an ML engineer. Would a Back-End engineer also work to become an ML engineer in this case? Would the constraint be it needs to use SQL and Python?
Your videos are eye-opening. thanks again.
Thanks. Exactly. A back end engineer is only good if your daily work includes working with SQL. You must have SQL skills on your resume. Most of your work will be sourcing data. That's all SQL in most real-world roles.
@@thedatajanitor9537 Thank you for the reply! I'll keep that in mind.
HI this video was really helpful and may I know if there are any entry level jobs in data engineering field straight out of college thank you
Nope. I haven't seen any.
@thedatajanitor9537 I have some SQL (over 10 years), programming (C# over 10 years and Python 1 year), No experience with Power BI and I want a remote job in Machine Learning, and I am outside the United States of America; Do I follow the Logikbot shortcut on Machine learning or the Data Analyst role. Please guide me.
Have you had remote jobs in the US before? If you haven't, getting a remote role is almost impossible. If you can get past a comprehensive SQL interview... you'll need to learn machine learning. LB will take care of the machine learning skills and certs but the remote angle will be really tough.
Hello Mike, I am a Phd graduate myself in ML. Would I also have to start with data analytics role before attempting to get into ML Engineer position? This is the position I am ultimately targeting at. Thank you for the videos btw.
No. As a PhD grad you should be able to land an ML role with your degree. You'll need to learn SQL but that shouldn't be difficult. A PhD at most companies will get you in the door.
Hello, very very helpful video sir, but I have a question. when I start with a data analyst role, won't I practice data cleansing well ?. so the only thing I need to do is to learn the modelling part ?
No. Data analyst don't cleanse data. Here, watch this.
ua-cam.com/video/Dfm67FlqCD4/v-deo.html
hello mister Mike, hope you're doing well. I want to ask you a question, do you consider 'prompt engineering' as one of the shortcut roles ? @@thedatajanitor9537
One question, I have the python professional certification, tableau data analyst certification, and excel associate certification. Where do you think I'd fit within my job search. I have no job experience and I really want to be a machine learning engineer. I do have machine learning skills but no certification yet
These are all worthless. All certificates are. Here you go.
ua-cam.com/video/PQo38qvbxy0/v-deo.html
Heh... I'm a SQL and BI veteran/real world data dude/machine learning engineer... and Kaggle competitions expert... SQL, especially SQL with vendor extensions like T-SQL - it's like the bass guitar. The bass is one of the easiest instruments to learn, but it is one of the hardest to *master*. SQL is exactly like that... now the SQL you see in training courses... versus the SQL you see and need to use in the real world of business data.. don't get me started. BI languages like DAX in Power BI and MDX are very very similar... easily to learn, so hard to master.
This guy knows exactly... and i mean exactly, what he is talking about. I don't think there's a video like it.
But my take is there is more work doing that stuff in the world of analytics and business intelligence, than there is in the world of machine learning. And just because you've done DA100, doesn't mean you are ready to model real Power BI solutions either :)
Thanks Josh. Much appreciated!!
I mean, yes, these job listings all ask for SQL and Python etc.. skill but the also ALL ask for Masters or PhD degrees as well. So i'm confused here. Is it or is it not, reasonably possible to transition from a pure python/sql programming job to a Machine Learning or AI etc.. position? This was my current goal but I'd rather not specialize in Python and SQL if it's impossible without a degree (I'm already past 30 and can't afford to work full time AND earn a master's with a Lung Tumor. I don't want to waste my time, but I'd like to be optimistic. If it was possible with an associate's or perhaps Bachelors i might be able to pull it off, but there's no way I could pull 6 to 8 years of university.
They ask for a lot but that's not what you need to have. No, I don't know a single machine learning engineer that knows their shit with a masters or PhD. Most of these types are theory kids with no real-world data skills. You won't get any data related job without a degree in something. The fastest path to a career as a MLE is via a data role.
I am doing my PHD in industrial engineering. Machine learning is probably going to be my research area. Do you think it’s going to be more beneficial for me to become a data scientist first and then switch to Machine Learning? I am fairly confident I can get a data scientist postion ( a lot of people in my school have). My end goal is machine learning engineer btw. Any advice.
You won't get a job in machine learning at any real company even with a PhD. Yes, I'd focus on getting a job as a faker scientist and then focus on learning the real-world skills you'll need for an MLE role.
@@thedatajanitor9537 how many years of experience would I need in a data science role?
Once you're in a research type role it won't take you long to move to machine learning. Six month or so.
@@thedatajanitor9537 By research role , do you mean anything in Data Science? So I can do a data science role for a year or so and then transition??
BTW your content is amazing. You're a scary dude but atleast you keep it real. I subscribed to your content.
Thanks for Awesome video, i am working as Senior Data Analyst (5 years XP) and wants to transition to MLE what would be the good learning course ?
My platform was made for this scenario. If you can get past the SQL interview and have 3-5 years of SQL on your resume then start on Logikbot.com.
Start learning Python and then data cleaning. Learning it is not taking the course and watching it like a movie. Learning it is taking notes and understanding damn near everything in the course.
is there any entry level Data Engineering (DE) jobs?, after about a month I will graduate with a master's degree in Artificial intelligence and Data Science, I saw some articles and videos about DE and it seems like a cool job where your work is impactful and helps companies to grow!
is DE a good choice?
Nope. A data engineering role is as hard to attain as a machine learning engineer role. You won't get any DE role without 3-5 years of hands on experience. I'm not trying to be mean but you're in for a long road ahead. A masters is worthless in the real-world these days.
Thank you!
As a high school student learning python and sql, should I pursue a masters degree in computer science to enter this job?
You should watch a lot of my video to get an understanding of the real-world. You need a bachelors and that's it. There are no entry level roles in machine learning. You have no chance with any degree at any level to get a job in applied AI. You'll need to start out at a more entry level job like a data analyst.
ua-cam.com/video/wlIKCJZEImw/v-deo.html
@@thedatajanitor9537 thank you for the reply, I’ll defo check out more of your videos.
How proficient would you have to be in python to truly have be successful in ML?
You need to know a good deal. I work with SQL and Python all day.
I'm partially in that role right now (as an Intern), and let me say that there is basically no other language used in this field outside of Python (small exceptions for R in limited cases and javascript if your company is legacy and fked up as all hell for production). To this day, outside of python, the ONLY other language I've used for ML is C++ just for hardware optimization on GPU stuff.
This is valuable knowledge! I was offered a free research based masters on top of a monthly stipend to do healthcare imaging segmentation work. I’m 25 and just graduated but I currently do web development. In your opinion can i do both as i get my masters or will the deep learning work require too much time?
Yep. You won't have a lot of free time but you can do both. It will pay off down the road but my suck for the next few years.
I appreciate your 2 cents. Not a lot of people around me who i can talk to about this stuff
@@sabinoramirez2600 Sure. Thanks for watching.
I am QA engineer from 2 and half year.. i have knowledge about SQL then JAVA selenium automation tool .. currently learning ML - thrue python .. i can get entry into ML ? Can i eligible for little short cut ? Please ans
Yep. You'll need to learn machine learning. You'll also need to polish those SQL and Python skills. If you think you can get by the Python and SQL interview questions now, start focusing on learning the ML process. If you aren't sure you can get past the Python here's a free course for you: ua-cam.com/video/foAMYMhwbsA/v-deo.html
@@thedatajanitor9537 thank you so much it's means a lot ..
@@yadnyeshgosavi7536 Sure. Thank you for watching
The path you propose for Masters and PhD's doesn't work. Companies won't even reply to our applications for analyst roles because they're afraid we'll move on quickly. We get interviews more easily for data science roles, but then interviews feel like we're facing an extreme bias against us.
I started programming at 8, was hand-assembling code to machine language at 11, then worked for few years as a programmer before college, where I used SQL a ton, then got a physics PhD, which was data analysis heavy, and then did a postdoc. I didn't go into academia because I don't feel up to it, I'm not that interested in teaching, and I could do my research nights and weekends.
In a functioning society, it should have been easy to find a job relevant to my skills, don't you think?
I looked for work in 2017-2019, failed, took a sabbatical to deal with the stress and heartache, and then started Doordashing during Covid. Now I'm on another sabbatical because I couldn't deal with the internal conflict that comes from Doordashing after a PhD and 2 years of failure at finding work. If it weren't for my fiance, I don't know where I'd be. My mental health, our relationship, and our long-term financial health have been ruined by this.
What's blocking the entrance ramp is the fact that employees are using companies as jobs programs for their friends. Apparently, no one gets a job without finding someone to pull them in the backdoor. My PhD adviser told me this, and a recruiter as well, but I didn't believe them, even though I had already seen evidence for it from the inside of companies. It felt like they were telling me everyone's acting like a rat on a sinking ship. I felt that surely, between my ability and humanity's intrinsic goodness, things would work out. Turns out they were right. I had too much faith in myself and humanity.
A portion of data science interviewers are lazy, superstitious, insecure, and downright stupid. They're looking for any half-baked reason to torpedo our applications, when the sum of what we bring would easily and obviously make us excellent hires in the intermediate term. Better hires than their college buddies, relatives, and interns. I've heard so much stupid shit from data science interviewers, like people with opinions they formed from N=1 samples, a guy at google who didn't know the geometric picture of how a perceptron layer works, a guy who told me I don't know what I want in life (don't tell me about what I want), and a guy who downright lied to his manager about an explanation I gave for gradient boosting.
Whether a person gets hired has more to do with connections than ability. That's a bad thing, not a good thing, but our society celebrates it.
If employers are worried about having to pay too much until we're up to speed, that's still on them, because as far as I've seen they don't want to actually have that conversation. In the early stages, they just want to know how much we're asking. They force it out of us ("It's just for this dumb form but we have to have it before moving on.") Then before we make it to actual negotiations, interviewers reject us (presumably because because we'd be too expensive and they already have someone they want to hire). It's a completely fucked way of going about it.
I don’t know about you, but my companies fought on Talent Service. They also sent very cool offers. Maybe you're just unlucky.
Why don’t you just get a web developer job of some sort like coding PHP and MySQL or go get some Microsoft certifications and try to get a junior .NET developer job… You might have to beg for a while but it’s much safer than risking your life doing shitty rideshare driving/delivery getting paid essentially nothing and risking your actual life from criminals/murderers and the road. You should also consider applying for some type of welfare benefits like SSI and SNAP food stamps too until you get a job. Then you can spend your time more productively studying LogikBot instead of dangerous driving.
If you are actually getting interviews, that seems like a sign that they consider you qualified at least. Maybe your interview skills are letting you down. Do you communicate well with people (especially less technical people)? Do you feel comfortable engaging with a stranger? These soft skills are very important in the recruitment process, even for technical roles. One type of soft skill is "telling your story", presenting all your overall professional and educational trajectory in a way that makes sense to someone else. I also come from a kind of weird background, so this was a common question I got, and knowing how to frame my experiences in a "coherent" way was important. Of course, it was all bullshit (I have a weird background because I in fact didn't know what I want - but that sounds bad).
I don't know what your salary expectations were, but maybe it is worth taking something low paying just to get your foot in the door. It can be easier to find a job when you already have a job. And even if you have to take a job that is "beneath you", you can still pursue some interesting projects in your own time to put on your resume or github.
I think it's also possible they are prejudiced at your PhD. Like, judging you for not being in academia or finding work, or judging you as being too expensive and overqualified. One potential tactic you might try is to undersell your resume. Don't mention the PhD, don't mention anything too esoteric and academic, just mention the hard skills like SQL, coding ability, data analysis, and tailor the resume for more modest analyst roles. Again, just to get your foot in the door so you can hop to a better company in a few months.
@@calohtar I agree. All good advice you gave him. Probably he needs to leave out the PhD when applying to programmer and analyst jobs.
This really isn't realistic.
"Please explain this four-and-a-half-year gap in your resume."
Umm...... drugs.
Best you can do is to give a plausible explanation for why you're leaving the field (e.g. not interested in a professorship) and hope they buy it.
I do not have a degree, but I believe I can easily learn any tech role. If I got the SQL, PowerBI and Tableu certa you mentioned. Will I immediately be rejected simply because I have no degree. I live in London
Yes. In the US and I'm guessing Europe a bachelors has become mandatory.
@@thedatajanitor9537 "less than 20% of people it IT around the world have comp sci degrees" according to your statement this means that the other 80% have a STEM degree, unrelated degree or no degree. This doesn't make any sense
@@MrPrebuttal What doesn't make any sense to you? What don't you understand? Yes, I'm saying 80% (much more actually) don't not have comp sci degrees and work in IT. I'm saying 80% of all the people in IT DO NOT have related degrees.
@@thedatajanitor9537 So 80% of candidates they hire have an unrelated degree, but the important thing is to have a degree. This is irrational.
@@thedatajanitor9537 This is why these shit companies do not innovate because you hire the lowest common denominators who are stupid enough to go to university and filter out more capable self taught candidates at the first stage of HR. By choosing an idiot who decided to get a degree in geography rather than somebody much more capable who is self taught.
☝🏼☝🏼☝🏼☝🏼
Hi Mike, I am working as PowerBI developer in a MNC from last 2 years and I recently got opportunity to move to a data engineering project in same company. Should I take it ?
My main motto is salary and I am ready to work hard.
Where do you want to end up? If you want to end up in data engineer or machine learning then hell yes. If you're a good PowerBi dev there are quite a few jobs out there paying really good money. Look for contractors opportunities on indeed.com and only look at one where the hourly rate is close to 80-100/hr.
@@thedatajanitor9537 i have previously worked on informatica as ETL developer but PowerBI interests me bcz i can see final end product with lot of analytical fun and looks easy to me than data engineering. But i am skeptical that PowerBI is easy tool and anyone can master in just few months. Whereas data engg is highly backend and more technical and need coding experience. What's your thoughts?
@@thedatajanitor9537 hey I am expecting your reply.
@@overlord7096 Hey. I don't work for you.
@@thedatajanitor9537 it's alright. But that was mean.
i am not able to pay for logikbot
can u provide razorpay or some other payment options for indians
I can do Paypal.
@@thedatajanitor9537 okay
how to proceed ?
@@machinelearning3518 Sorry. Just email at mw2398@hotmail.com
I am a business analyst with over 7 years experience. I am very proficient with SQL, Python, data visualization tools, etc. I even taught myself machine learning and applied that at my current job (I created and validated an accurate predictive model and implemented in an application). Still can't get a data analyst job. Much less a data scientist or machine learning job. I have been searching for a while. Is it just the job market?
It's you. Sorry Justin. There are tons of jobs out there. Are you not getting interviews are not getting past them? You'll need to focus on the data analyst role at first. It's in line with what you are doing and you're experience as a BA will carry weight? What tool are you accomplished with? PowerBI, Tableau or Looker? What certifications do you have? NOT completion certificates!!
@@thedatajanitor9537 I appreciate the response! In the past 1000 applications I have had three interviews (not counting recruiter screens) and felt they went great (one of them went the entire process). The majority of them are data analyst roles. My resume has been refined and tailored for most (followed guides, had it reviewed, etc). As a BA I am the lead data analyst and have done a lot of work with data using Excel, SQL, Power BI, and Python. I have a bachelors in Mathematics, minor in Computer Science, no other certifications (I do not believe they help get an interview but I am not opposed to getting one).
@@shermin2o9 That's very strange. Everything sounds in order. Certifications are a big deal at every company I've worked for. At Uber, we only interviewed those with top certs.
I'd suggest the PowerBI cert. Put it at the top of your resume. Are you applying for remote roles also?
@@thedatajanitor9537 I will definitely look into the certifications, thanks for that insight. And yes I am applying for remote roles only. Appreciate the advice!
@@shermin2o9 yo this is scaring me, a ba in math and 7 years experience as a data analyst O:? so then mike the datajanitors advice is useless then? or have you been able to get a job since this reply? was it the powerbi cert?
I'm mostly interested in creative challenges like developing new tools that serve practical purposes like what you see on channels like Two Minute Papers. Do you know how people get into those sorts of jobs? I like to think that if I have lots of personal projects showing my problem solving ability and practical creativity towards developing new technologies that I could get a job somewhere, but I have a feeling that it doesn't work like that.
I do this for fun and am not seriously looking for a career change, but I was wondering if you knew much about this? Nividia and google do quite a bit of ML technology development and that's what I'm most interested in. How did those people get their jobs, and are the jobs as glamorous as they seem?
Thanks
Not exactly sure what you are asking. Who writes machine learning models from scratch? Hardcore programmers at Google, Microsoft... etc. That's about it.
Here's a course I have on authoring models from scratch. Again, this is brutally difficult. It requires sold math skills and more importantly, really hardcore programing skills most do not have.
www.logikbot.com/courses/building-machine-learning-algorithms-in-python
So, I think I answered your question. The only people authoring the top machine learning models like Tensorflow or Keras work in big tech. They are elite C and Python programmers. That's about it.
It's brutally hard to get these roles.
@@thedatajanitor9537 thanks!
@@cory99998 Sure.
Hi Mike, I know this comment is abit late but right now Im going to take a masters in Data Sc at Monash Australia so that I will be able to migrate there quicky, but I understand what you said about the skills in masters to become a machine learning engineer is hard, so would you recommend me to take an external SQL courses on top of my masters so that when I graduate I can start immediately with knowledge in SQL?
Yes Darren. You should be learning SQL and Python and applied machine learning right now. A job in applied machine learning without SQL on your resume these days is almost impossible.
@@thedatajanitor9537 Alright, will do! Currently learning python again after attending a comp sci fundamental course in University, but I will make sure I add SQL asap! Thanks for the response!
To be honest. Machine Learning feels like over hyped garbage algorithms. In bigger countries its useful but in smaller countries like NZ it is such a waste of time. Our data isn't even clean enough to make any of the cool data shit work well. 😒
It's embarrassing.
Yesterday, I got laid off from my first-ever job as a machine learning engineer. I was there for all of 8 months. I was the only person at the company with the job title of machine learning engineer. From my very limited experience, much of what you say rings true to me. I graduated last year with a master's degree in data science after having having worked the first 50 years of my life in other fields. I began the degree program with zero experience in Python or SQL, and accepted a lower salary as a machine learning engineer because of my relative lack of experience. I struggled mightily, and will take to heart your suggestions as I look for my next job. I don't regret getting my degree and I don't regret my first machine learning job. I love learning stuff and I'll be checking out what you offer.
Please, do not be over enthusiastic. The complexity in this career and how messy are the software, will make you quit at some point.
only truth in this video, a PhD won't cut it in the real world nobody cares
They don't. I've worked with tons of them. They are moved to the business side of the house in three months or less. Here's a good one.
ua-cam.com/video/jOg_0OfMHhU/v-deo.html