This is surely great Devendra sir from the perspective of complete resource guide for the career. It'll be truly helpful in a greater way if you could provide with some sequence of learning (like a roadmap) to follow for the foundational buildup.
The delivery seems fine to me. Maybe the audio could be made more clear. That being said, it will be helpful if you could upload some demo videos on ML.
upload video on the machine learning algorithms from the basic to advanced not cover any RU-vid video internet that level with your experience its will helpful
Introduction to Probability by Dimitri Bertsekas is a textbook used at many many universities. All of the statistics by Larry Wasserman for stats. Any good ML book also covers a bit of probability and statistics, like Probabilistic Machine Learning or Introduction to Statistical Learning or Elements of Statistical Learning.
@@mohammedaltaf4316 Arpit Bhayani has a course on system design for beginners and a different course for intermediate level. Also MLE roles have system design rounds. AS have ML design rounds. RU-vid is always an option. There are few good channels on RU-vid but can't remember right now.
Update: its 22k INR at thecuriouscurator.in and official start date is 1st March. It is a mix of recorded videos and weekly sync up for clarifications. We will also have monthly AMA sessions related to career guidance.
Thank you for all your videos. Please do keep up with them. I am learning a lot. I am completing my bachelors as a statistics specialisation student in data science and machine learning. And I am slightly clueless about proceeding and cracking the entry level data science interviews. Please do make video on that.
The curriculum looks detailed, but I was hoping you would add the following in the course: 1. Interview questions for every module & submodule. 2. System Design for Machine Learning. 3. A few end-to-end ML/DL projects that's worth putting on the resume while interviewing for FAANG. Apart from that, I have no complaints regarding the syllabus, and very excited to join your course!
Interesting, but I have the following queries: 1. Can you please give a tentative list of topics covered in the course? A Google doc of the detailed list of topics would be helpful. 2. What's the time distributions for each of those modules? 3. What's the end goal? Are you training the students to become a MLE, or is this geared towards a research career, like that of an Applied Scientist at FAANG? 4. Since, there are a bunch of other courses (paid + free), how is your course any different? Personally, I would love a good balance of both practical, and theoretical understanding. I'm looking to interview for MLE, as well as Applied Scientist roles in 1-2 years time, and so I'm very interested in your course. Looking forward to hearing further details from you! Thanks.
I will get back soon with a doc. Regarding MLE this is more than enough for ML part but it also has an engineering part (like DSA leetcode style and system design). For applied scientist more work should be required like reading research papers etc but i should be able to provide guidance in terms of expectations (we will cover some good research papers because they opened my mind so I hope it will give insights to everyone). The goal of course is supposed to be for future applied Scientist. We will definitely cover a lot of stuff, few things which if it comes up will impress the interviewers. But also 1 year is such a short amount of time for beginner to FANG. I will do my best but it depends on many factors including motivation of students.
@@theCuriousCuratorML Ah, I see. Right now, I'm honing my algorithmic skills by reading CLRS, and solving lots of problems from platforms such as LeetCode, and GeeksForGeeks. In total, I have solved roughly 300 problems so far, but I do understand that problem count is irrelevant, and it's the algorithmic design strategies that matter the most. Hopefully, I will be interview-ready in this aspect by the end of the year. I'm extremely happy to hear that we will be discussing research papers on this course, since I'm very interested in fundamental ML research, and aiming for Applied Scientist roles, and other research-intensive roles at FAANG. In the past, I have tried implementing classical papers such as t-SNE, and UMap, but always felt that I could do more with the guidance of an expert like you. 1 year indeed sounds a little short, but I can assure you that there won't be any lack of efforts from my side. Nonetheless, I do believe that extending the course to, say 15 months, would be better, since it could take a while to fully understand and appreciate the material that's being covered in the course. Thanks again for taking the time to read my queries, and I hope to hear from you soon!
@@tirthasg Thanks for your comment, I have the same plan and as I was reading your comment, I was feeling like you are reading my mind. :) I guess you are part of the course now ?
The moment He said said after all this 6+ months of internship and then he got to knew that he knew nothing. AND here I am with a plan to become applied scientist in 8 Months🥲. Aree Koi hai Bachaa Lo Re. 🥲
Usually a PhD is more than enough but which field or which group also matters. But ofcourse there are specific groups in companies like deepmind which have their own requirements. A PhD doesn't guarantee anything and it's easier task to get a job in industry. You need to polish or aquire skills needed in industry. A PhD in machine learning and related field will usually give you the skills but you should not forget practical aspects because PhD can also be very theoretical.
@@theCuriousCuratorML People said that You need Phd to get Research scientist role in big tech....What are the maths require to become Research scientist in computer vision? What would be your roadmap of doing computer vision?
@@convolutionalnn2582 if you want to be research scientist at let's say Google or microsoft research. You need stellar publication record. And with publication in top tier conference you will also learn the background needed while on your PhD journey. Tensor algebra and linear algebra. Analysis and optimization. Machine learning. Probability and statistics. Computer vision, multiple view geometry. Calculus. These are few of the subjects you need to be familiar with. Depending on research you might also get expertise in certain subfields. Also i am an applied scientist and that requires more practical skills and knowledge. It's ok not to have research papers in applied science position in industry.
@@TheCuriousCurator-Hindi I am gonna do integrated phd in 4 years.....Any advice on my 1st years(Master) or 2 or 3 or 4? In each year what should i focus on?
Reading all of this is really interesting, but you forget once you read, you've to revise and all, i prefer making my own notes for whatever topics i need to study, so theyre a bit short, how do you tackle this problem of forgetting stuff ? Anyway nice collection of books lol
Never miss an opportunity to brainstorm with a colleague. Have been doing it for a while. I also revisit and cover newer topics once every few years. Reading same thing again after few years gives a new perspective because you have also grown since you last looked at it.
@@theCuriousCuratorML Why do you recommend those heavy math books ? I actually will take those courses in the first semester in my masters but i dont see questions from them being asked in interviews
@@sanjotsagar1458 they may not but we don't work in silos. Having some understanding of ml ops helps you communicate better. Also it's very company specific and sometimes a part of it needs to be done by you. Another reason is curiosity plus industry trends. Employers needs super human abilities. You should be good at everything 🤣 You don't write poetry but they made you read them in school. Similarly we need to have our fundamentals and we may not use all aspects of it. Some interviewers can ask something from it. Maybe it's not math heavy, may be it is.
List of resources for Applied Scientist Career. 1. www.algoexpert.io (bundle pack 95$) or leetcode (50 easy 70 medium and 30 hard problems) 2. Probabilistic Machine Learning: An Introduction by Kevin Murphy 3. Probabilistic Machine Learning: Advanced Topics by Kevin Murphy 4. Machine Learning a Probabilistic Perspective by Kevin Murphy 5. Algorithms for Optimization by Mykel Kochenderfer and Tim Wheeler 6. CS229 Machine Learning lecture notes at SEE (Stanford Engineering Everywhere) 7. Linear Algebra by Gilbert Strang 8. basic probability and statistics Reading list this year: 8. Trustworthy online controlled experiments by Ron Kohavi, Diane Tang and Ya Xu 9. The System Design Interviews by Lewis Lin 10. Building Data Science Applications with FastAPI 11. Practical ML Ops by Noah Gift & Alfredo Deza 12. Engineering MLOps by Emmanuel Raj 13. Designing Data-Intensive Applications by Martin Kleppmann 14. Bandit Algorithms by Tor Lattimore Optional: 15. Convex Optimization Theory By Dimitri Bertsekas 16. Counterfactual and causal inference by Morgan and Winship 17. Dynamic Programming and Optimal Control Volume 1 and Volume 2. 18. Linear Algebra and Learning from data by gilbert strang
TLDR: For applied research there is no requirements on degree. For well paid academic type research jobs (say FAIR or Google deep mind or Google brain, microsoft research) PhD is needed. For applied research, Master's is highly recommended but from my limited experience there is no restriction on degree. But usually it is correlated with knowledge. I have never met a fresher bachelor's student who is good at ML. How can they be.. without spending 1000+ hours reading a bit here and there. If you want a job where you publish papers masters can be required but having good papers in good conference will get you the job. Top research jobs pay well in this space but does require a PhD. Usually world class.
Hi, I wanted to know what is the scope of Applied Optimization and Parallel Scientific Computing jobs in private companies in India and abroad. Can you pls help me with it
This is a difficult question. Probably operations research roles will be closest high volume jobs to applied optimization. Machine learning research roles can also be considered in the same category. Parallel computing and scientific computing probably the research roles at Intel or similar. Number of jobs will be less so in a way companies will be very selective. Probably look for a PhD.
Hi sir. Amazing video.Thanks a lot. I need your advice on Research Engineer/ Applied Engineer positions prep. Is there any way to contact you on email ? My email id is : amitab12690@gmail.com . Thanks a lot once again
If you are serious about ML this is not the only book you will ever read. Just get started. ESL is also a good book. Keep learning new algorithms and paradigms this book or other books. Explore more topics by whatever resources you find good for yourself.
I am in my final year of Masters in Computer Application. Just last year I realized that I don't want to go in software development jobs, but I want to go for research. I have worked on some ML projects that involved image processing. Then I also completed some deep learning certifications from courtsera. How do you suggest I should move forward?
Deep learning is hot. I guess do some project of decent complexity and get a job/internship in a similar space where your strengths lie. Job or internship make sure you are paid a decent pay. Read a lot. The video gives a glimpse of breadth you can choose from. Read research papers in some area of your interest. It's also a demanding space, you have to add value to your company and be updated as well. Simpler things will be automated soon. Also no one pays you to study. That makes it tricky prioritisation.
Follow this channel. I will create more content when I can find time. I have a playlist on fundamentals called yet another tutorial on linear regression. Don't be confused by the simplicity of the name.
Hello, this is kinda embarrassing but I have a question,idk who should I ask so anyone seeing this can answer or give suggestions, I graduated in 2018 and have no job exp, been preparing for govt exams but couldn't get through. The selection process is really long, for eg, in some cases, they can drag it up to 3 years from the date of adv to date of joining. I did get campus placement offers but I didn't like the profiles at that time and the ones I wanted... I wasn't good enough for them. My % is in 60s. But I want to go back to my roots and join a reputed firm. I'm willing to put in hard work. So should I prepare for FAANGM? I've no job exp and this gap is gonna keep increasing. Even if there's a slight chance of me being selected I'll put in hours but if there's none then I don't want to waste my time and hinder my govt exam preparation. I've lost contact with almost everyone and people here seem to be in touch with firms or recruitment processes so that's why I'm asking. In short, I've done nothing for the past 2.5 years after graduation and if I start coding and practicing rn can I get into any good firm in 2021?
You can but you are underestimating the work required. It might be comparable to the time it takes to crack a good exam. Atleast for fang companies. That said, it is possible but very difficult. My advice is that don't be too rigid in life. Take feedback and learn what works for you. If you cannot get into fang now maybe with some experience you can. Take a decent job if you get it. Best wishes.
Sir, I am learning data science from simplilearn. It is a master program. I do not have a bachelor's degree. I am only 12th pass with science (math) and working as a medical transcriptionist for the last 12 years. I am from India and my aim is to get job in USA. Plz guide me.
This is not something I am good at. I might know how to do it for someone in my situation but I don't know how to do it in general. One way might be get into a top technology company. But I don't know if big companies will hire someone with your credentials for tech role. I have heard Google doesn't need a degree. That said and assuming you get a job there then work for few years there. Gain experience. Internal transfers in companies like Amazon or Google is easy. That can enable you to move to US. Directly getting a job there and someone sponsoring your visa might be difficult.