How do I get a machine learning job after 6 months of practice?

Updated on : December 6, 2021 by Freddie Powell



How do I get a machine learning job after 6 months of practice?

On Telegram there is an AI job group name that you must have to join that group. They publish daily papers on artificial intelligence, machine learning, data science.

I do not have a CS degree. I just took an introduction to computer science class. I started with ML earlier this year and have several interviews with companies that contacted me through Quora. In fact, one of them is in 14 hours.

The truth is, they stop caring what grade you have and where you got it the moment you walk through the door. The diploma itself is nothing more than a tool to enter through that door. Having a CS degree simply tells a company that it may not be a total waste of interview time. If you don't have the title, you just have to show it in another way.

For me, that's the way it was

Keep reading

I do not have a CS degree. I just took an introduction to computer science class. I started with ML earlier this year and have several interviews with companies that contacted me through Quora. In fact, one of them is in 14 hours.

The truth is, they stop caring what grade you have and where you got it the moment you walk through the door. The diploma itself is nothing more than a tool to enter through that door. Having a CS degree simply tells a company that it may not be a total waste of interview time. If you don't have the title, you just have to show it in another way.

For me, that's the way it was answering questions about Quora. Not why I'm answering on Quora, but it's a nice bonus. However, it doesn't have to be the same for you. You can do anything as long as you show that you are experienced, competent in this field to the point where they could benefit from your skills.

The biggest difficulty is that you usually have to do it in the form of a resume. That is a short document that will probably only be read. It is better to inform them of your existence before they even know if you have a CS degree or not. Make an interesting project and put it on GitHub. Better yet, contribute to the open source libraries. Start your own open source project. Anything that shows "I know how to do this."

If you want to be a bit more ... uh ... I guess original is the word, you could do something like this: train a convolutional style transfer network to basically recreate an image by typing "hire (your name here)" all over the place. Then put it at the top of your resume. However, make sure the results look really good. Your goal here is simply to get them to read a little more carefully. After that you can describe how you did the style transfer net.

That approach will be pretty hit or miss I guess, but all you really need is a few hits.

The answer is emphatically yes.

Machine learning jobs, in general, have four different types:

  • Data theorists are the people who research academic concepts related to artificial intelligence and create algorithms such as Deep Learning or Calibrated Quantum Mesh. These are typically PhDs and are generally off limits for non-academic or non-R&D candidates.
  • Data architects are the people who take algorithms from data theorists and apply them for real-world use cases. For example, use CNN for facial recognition. You can get to an architect by being a data analyst / modeler, but it is unlikely that
Keep reading

The answer is emphatically yes.

Machine learning jobs, in general, have four different types:

  • Data theorists are the people who research academic concepts related to artificial intelligence and create algorithms such as Deep Learning or Calibrated Quantum Mesh. These are typically PhDs and are generally off limits for non-academic or non-R&D candidates.
  • Data architects are the people who take algorithms from data theorists and apply them for real-world use cases. For example, use CNN for facial recognition. You can become an architect by being a data analyst / modeler, but it is unlikely that you will start your machine learning career here. An exception is if you are already working in a data-driven domain and you know more about it than any data scientist. In that case, you should try to join a data science team that pairs you with ML experts and teaches you through the process.
  • Data modelers are the people who apply the data architects model to real-world business decisions and maintain the process. For example, the Facebook team that is implementing and scaling its facial recognition capabilities. This could be a good place to start. Facebook still prefers PhDs for these roles, just because it can, but there are plenty of other teams that just need smart people with basic ML skills - the kind of knowledge you can get from MOOCs and universities.
  • Data analysts are the people who find and clean data so that data science models can run. The skills required are perseverance, resourcefulness, some statistical orientation, and being good at your jobs. This is another great starting point for a career in machine learning. I know people who started out as data analysts, continued to grow their skills, and have now become data architects.
  • Full-stack data scientists is a term used by data science teams when they don't know exactly what they want in their data scientists. I will say that these teams are usually bad news with two exceptions: 1. If they are too small, for example, a startup, and cannot afford individual roles; or 2. they just don't know what their needs are and just want to start somewhere. In any case, while looking for full-stack data scientists, they would actually have room for experts and beginners. If they can get in, this might be the best way to teach themselves machine learning!

I hope this helps.

It is not that difficult if you have prepared well. For multinational companies with superior products, preparation must be more rigorous, which can make the task difficult. Another aspect is that who you are competing with, sometimes they are people who have spent a couple of years learning to achieve that rigor. So you have to spend comparable time / effort here (compared to people who only fill these roles). Spending a lot of money on online courses will guarantee nothing, there is quality free content out there. I have seen courses that are sold made by people who themselves will not pass scientific research interviews in the best professionals.

Keep reading

It is not that difficult if you have prepared well. For multinational companies with superior products, preparation must be more rigorous, which can make the task difficult. Another aspect is that who you are competing with, sometimes they are people who have spent a couple of years learning to achieve that rigor. So you have to spend comparable time / effort here (compared to people who only fill these roles). Spending a lot of money on online courses will guarantee nothing, there is quality free content out there. I have seen sold courses made by people who themselves will not pass scientific research interviews at major multinational product companies. If you want to spend on courses, at least follow someone who has the ability to pass these interviews. Some channels are popular simply because they skip the depth and make you think it's not much. Follow the best colleges' classroom lectures on YouTube and not some guy who doesn't know what he's talking about. If you spend enough time learning, there are enough free things that are enough to get you started in the industry.

Read real ML and related books, run complex projects, write research reports / articles, follow the research community, and practice data structures and algorithms. This is something that people do to get the best jobs in the machine learning space.

If you're wondering why I should try so hard, here's a motivation. Good machine learning scientists get paid very well. After 4-5 years in the position in major multinational product companies, you may get paid north of INR 50-60 LPA in India. This number is pre-tax income numbers, as opposed to those shown at college. In college, you should always ask what your pre-tax income is in year 1.

Another aspect is that you will like to do the job that you are good at. I'm not saying that the job doesn't demand a lot of skills. Yes, it requires a lot of theoretical and practical skills. But it might be worth it for someone who is passionate about this field and ready to have the fundamentals that will allow them to learn with a job forever. Don't think you will have an incredible amount of time to learn something. And there are topics that you have never heard of and without solid foundations it will take months. It cannot be assumed that there is a library for everything.

Introduction

Not so long ago, using the pivot table option in Excel was the upper limit of my number skills and the word python was more likely to make me think of a dense jungle or nature show on television than a tool to generate business information and create complex solutions.

It took me ten months to leave that life behind and begin to feel like I belonged to the exclusive world of people who can distinguish their medians from their media, their neighborhood pub x bars, and who know how to teach machines what they need. learn.

The transformation process was not easy and required har

Keep reading

Introduction

Not so long ago, using the pivot table option in Excel was the upper limit of my number skills and the word python was more likely to make me think of a dense jungle or nature show on television than a tool to generate business information and create complex solutions.

It took me ten months to leave that life behind and begin to feel like I belonged to the exclusive world of people who can distinguish their medians from their media, their neighborhood pub x bars, and who know how to teach machines what they need. learn.

The transformation process was not easy and required a lot of work, a lot of time, dedication and a lot of help along the way. It also involved more than hundreds of hours of "study" in different ways and an equal amount of time practicing and applying everything that was being learned. In short, it wasn't easy to go from being a data sucker to a data nerd, but I did it while going through an insanely busy work schedule as well as being the parent of a one-year-old.

The goal of this article is to help you if you are looking to perform a similar transformation but don't know where to start and how to proceed from one step to the next. If you're interested in finding out, read on to get an idea of ​​the topics you need to cover, and also develop an understanding of the level of expertise you need to develop at each stage of the learning process.

There are many great online and offline resources to help you master each of these steps, but very often the problem for the uninitiated can be figuring out where to start and where to end. I hope spending the next ten to fifteen minutes reading this article will help you solve that problem.

And finally, before continuing, I would like to point out that I had a lot of help to make this transformation. Right at the end of the article, I'll reveal how I managed to squeeze out so much learning and work in a matter of ten months. But that's for later.

For now, I want to give you more details about the nine steps I had to go through in my transformation process.

Step 1: understand the basics

Take a couple of weeks to improve your "general knowledge" about the field of data science and machine learning. You may already have ideas and some understanding of what the field is, but if you want to become an expert, you need to understand the finer details to a point where you can explain it in simple terms to just about anyone.

Suggested Topics:

  • What is Analytics?
  • What is data science?
  • What is Big Data?
  • What is machine learning?
  • What is artificial intelligence?
  • How are the above domains different and related to each other?
  • How are all the above domains being applied in the real world?

Exercise to show you know:

  • Write a blog post that tells readers how to answer these questions if asked in an interview.

Step 2: learn some statistics

I have a confession to make. Although I feel like a machine learning expert, I don't think I have any level of expertise in statistics. Which should be good news for people who struggle with concepts in statistics as much as I do, as it shows that you can be a data scientist without being a statistician. With that said, you can't ignore statistical concepts - not in machine learning and data science!

So what you need to do is understand certain concepts and know when they can be applied or used. If you can also fully understand the theory behind these concepts, give yourself a good pat on the back.

Suggested Topics:

  • Data structures, variables, and summaries
  • Sampling
  • The Basic Principles of Probability
  • Random variable distributions
  • Inference for numerical and categorical data
  • Linear, multiple and logistic regression

Suggested exercise to mark the completion of this step:

  • Create a list of references with the easiest explanation you found for each topic and post them on a blog. Add a list of questions related to statistics that you are expected to answer in a data science interview

Step 3: learn Python or R (or both) for data analysis

Programming turned out to be easier to learn, more fun, and more rewarding in terms of the things it made possible than I had ever imagined. While mastering a programming language could be an eternal quest, at this stage, you need to become familiar with the process of learning a language and that is not too difficult.

Both Python and R are very popular, and mastering one can make the other easier to learn. I started with R and gradually started using Python to do similar tasks as well.

Suggested Topics:

  • Supported data structures
  • Read, import or export data
  • Data quality analysis
  • Data cleaning and preparation
  • Data manipulation, for example, sorting, filtering, aggregation and other functions.
  • Data visualization

Know that you are ready for the next step:

  • Extract a table from a website, modify it to calculate new variables, and create graphs that summarize the data

Step 4: Complete an exploratory data analysis project

In the first cricket test match played (see scorecard), Australian Charles Bannerman scored 67.35% (165 out of 245) of his team's total score, in the first innings of cricket history. This remains a record in cricket as of this writing, for the highest share of a batter's total score in an inning of a test match.

What makes the innings even more remarkable is that the other 43 innings in that test match averaged just 10.8 runs per inning, with only about 40% of all hitters recording a score of ten or more runs. In fact, the second highest score by an Australian in the match was 20 runs. Given that Australia won the match by 45 runs, we can say with conviction that Bannerman's innings were the most important contributor to Australia's victory.

Just like we were able to build this story from the test game scorecard, exploratory data analysis is about studying the data to understand the story behind it and then sharing the story with everyone.

Personally, I find this phase of a data project to be the most interesting, which is good, as you might expect much of the time in a typical project to be taken up by exploratory data analysis.

Topics to be covered:

  • Single Variable Scans
  • Paired and multivariate examinations
  • Viewing, Dashboard, and Storytelling in Tableau

Project exit:

Create a blog post that summarizes the exercise and share the board or story. Use a dataset with at least ten columns and a few thousand records

Step 5: create unsupervised learning models

Let's say we have data for every country in the world on many parameters ranging from population to income, health, major industries, and more. Now suppose that we want to find out which countries are similar to each other in all these parameters. How do we do this, when we have to compare each country with all the others, on more than 50 different parameters?

That's where unsupervised machine learning algorithms come in. This is not the time to get bored with details about what they are about, but the good news is that once you reach this stage, you will have entered the world of machine learning and are already in elite company.

Topics to be covered:

  • K-means grouping
  • association rules

I finished the Machine Learning Nanodegree this September. I joined a gaming company (Sports Port Gaming Studio) as an algorithm analyst. Although my job description specifies that I work with algorithms, I am also responsible for developing bots using Reinforcement Learning to make users play in the game. I am very excited to be able to apply what I learned at ND. I also know some of my colleagues who work with Machine Learning after their DN. One of them working on backend systems became the machine learning guy at his workplace after his ND. Use and apply what you learned in

Keep reading

I finished the Machine Learning Nanodegree this September. I joined a gaming company (Sports Port Gaming Studio) as an algorithm analyst. Although my job description specifies that I work with algorithms, I am also responsible for developing bots using Reinforcement Learning to make users play in the game. I am very excited to be able to apply what I learned at ND. I also know some of my colleagues who work with Machine Learning after their DN. One of them working on backend systems became the machine learning guy at his workplace after his ND. Use and apply what you learned in the ND on a daily basis. Another student I know is working as a machine learning intern at Intel. He was offered the opportunity even before he finished his ND. It's more,

I think the opportunities are many, as the field is nascent and by the time you complete a Master's degree in the field, the demand will be down. The ND is pragmatic in today's ever-evolving job market and can show potential employees a ton of projects.

Sure. It's possible. Case in point: my brother.

My brother is smart, but he lacked direction and never finished his studies, after changing the subject several times. He was then off duty with some social phobia for a few years. After he was diagnosed with that, he began to return to the world. He got a clue about an internship that he thought sounded interesting. It was with a company that uses the Unity game engine to create virtual reality environments for city planning and construction projects. For that I needed to know how to program in Unity. I've done some recreational programming,

Keep reading

Sure. It's possible. Case in point: my brother.

My brother is smart, but he lacked direction and never finished his studies, after changing the subject several times. He was then off duty with some social phobia for a few years. After he was diagnosed with that, he began to return to the world. He got a clue about an internship that he thought sounded interesting. It was with a company that uses the Unity game engine to create virtual reality environments for city planning and construction projects. For that I needed to know how to program in Unity. I've done a bit of recreational programming, so he asked me to help him figure it out. I had a few weeks to learn enough to impress them. So we started playing with him and we did a nice little game.

This was enough to get him hired as an intern and a few months later as a full-fledged employee. Even when he was still technically just an intern, he was already leading a team. Now, a year after we started playing with him, he is conducting job interviews, giving talks, and other companies have started looking for him.

However, we never quite finished that game. Now you are doing more important things.

In theory, there's a chance that you could get a job in the ML field even without studying ML at all ... that's how probability works. However, according to expected utility theory 1 (for a rational agent), an agent must take those actions that maximize its general utility in the long run. Getting a job is never easy for the wrong person, but it is not difficult for the right person either. Look for ML job postings on the Internet or on company websites that hire ML professionals. You can find the answer yourself. Look specifically: What are your technical requirements? What skills are they looking for

Keep reading

Footnotes

1 Expected utility hypothesis

In theory, there's a chance that you could get a job in the ML field even without studying ML at all ... that's how probability works. However, according to expected utility theory 1 (for a rational agent), an agent must take those actions that will maximize its overall utility in the long run. Getting a job is never easy for the wrong person, but it is not difficult for the right person either. Look for ML job postings on the Internet or on company websites that hire ML professionals. You can find the answer yourself. Look specifically: What are your technical requirements? What skills are they looking for? What are your educational requirements?
Many people who can say YES are those who have no ML education and have not earned their Udacity degree, so take each tip with a spoon (without a hint of it) of salt. Good luck and keep learning.

Footnotes

1 Expected utility hypothesis

Yes, if you are a T-shaped person. If you have strong problem-solving skills, some knowledge of data, along with some knowledge of statistics before you start learning ML, it is possible. T-shaped people can cope with most of the things you throw at them. I have witnessed this happen a few times where people without very strong ML knowledge were assigned to submit some existing model that runs on CUSTOM data, and they were tasked with collecting the data, cleaning and preparing the data, evaluate the model and put it into the existing cloud service. Those people had excellent problem-solving skills, good data from ba

Keep reading

Yes, if you are a T-shaped person. If you have strong problem-solving skills, some knowledge of data, along with some knowledge of statistics before you start learning ML, it is possible. T-shaped people can cope with most of the things you throw at them. I have witnessed this happen a few times where people without very strong ML knowledge were assigned to submit some existing model that runs on CUSTOM data, and they were tasked with collecting the data, cleaning and preparing the data, evaluate the model and put it into the existing cloud service. Those people had outstanding problem-solving skills, good data backgrounds, and knowledge of how to process large amounts of data. They did it in less than 2 years.

Today, machine learning is being used in all aspects of the field such as IT industry, e-commerce finance, business, banking, education, and even governments are using machine learning.

Mainly ML, it is being used in IT, e-commerce and banking (international banks).

Therefore, both machine learning and deep learning are at the peak of high demand. It is a very good career option.

The incline of the machine is one of the most demanded races lately. Therefore, pursuing a career in machine learning is a good idea if you like coding.

If you want to know how to learn machine learning, check out this blog: Shiv Raj Singh's answer to What are the resources or courses to get started with machine learning?

In this blog you can find the best way to learn machine learning for beginners.

Happy coding.

Other Guides:


GET SPECIAL OFFER FROM OUR PARTNER.