Metis

This is my review of METIS Data Science (DS) Bootcamp experience in Seattle, winter session, Jan-Mar 2019.    I really enjoyed my bootcamp experience! Everyday I was coding daily, learning new concepts, and implementing machine learning (ML) algorithms into various projects pertaining to my interests. It was a blast!!!   Some things that I really enjoyed: (1) The format emphasizes creating design-to-product pipelines for data science projects. Having done 4 independent projects, I felt ready to tackle new problems, (2) my instructors, Chad and Cliff, provided a good mix of academic and pragmatic approaches to learning data science and completing DS projects, (3) after completing the bootcamp and reviewing class material for a couple weeks, I felt ready for interviews, (4) we practiced pair programming on a daily basis, (5) I felt a comaraderie spirit with my peers. (6) in terms of career support, our counselor (Marybeth) was very helpful. She helped me with everything I needed, from resume/cover letter review, job hunting strategy, connecting with other recruiters, etc.    Some things to improve (perhaps only for my cohort): (1) having code reviews or detailed project reviews with students would help I think, (3) I would’ve wanted to have more whiteboarding practice (this one didnt come up until the end, during mock interviews), (3) learning more DS use cases for various business problems would’ve really helped.       Advise for future students: 

• The bootcamp provides sufficient (elementary) overview of fundamental machine learning algorithms, at least enough for entry-level data jobs.  However, I’d advise future students to also spend some time learning from external resources (books, blogs, etc.) to get a deeper understanding of each concept. 
• If you intend to join the bootcamp, look up the curriculum ahead of time and think about what kinds of projects you might want to tackle based on your interests. 
• Unless you already have previous programming experience, it would be good to learn python programming and other CS fundamentals yourself.
• Just keep in mind that career transitioning is not as easy as you might imagine. If you are trying/planning to do it, I think your quickest way to getting a DS role is to leverage the domain knowledge you have from your previous career and then implement DS/ML into your field. Also, I learned that becoming a data scientist is more like a marathon, rather than a sprint. 

I was impressed by how well the course was designed for people at many levels. Since it is a project-based curriculum, you were free to bite off whatever you felt like you could chew. For example, those who were CS majors could dive a little more deeply into new packages/techniques since the coding wasn't as much of a hurdle, whereas those with less coding experience could stick with what we discussed in lecture. I loved the balance of project time and lecture time and felt like it was the best way I've ever been able to learn. 

It is fast paced, and to grasp everything that is discussed would be a challenge for almost anyone. However, you get a great understanding of the topics/techniques that you choose to use in your projects. Interviewers know you wont know everything, and simply want to see that you have a deep knowledge of techniques that you actually implemented. 

I learned a lot from my classmates. Everyone was always willing to help others if they felt like they had more experiences with a topic. I think the admissions process selects for people they think will be successful working data scientists, which includes traits of personability and teamwork.

The careers team was so knowledgeable and supportive. The instructors were so smart and invested in your learning. I learned more in these 3 months than I did in any year of college (technically and about networking/job search/careers) and I enjoyed it much more. 

I wouldn't have been able to revamp my career without Metis. They were able to teach alot of different concepts fairly in depth (within reason for a 12 week program) and their focus on 5 projects I believe is the difference maker for Metis. It allows you to show employees you are relatively ready to perform well for their company. The extra job assistance post Metis was better than what my University had to offer and I ended up with a solid job after Metis. I highly recommend this program to anyone who is trying to break into the Data Science industry.

Overall, I'm glad I did Metis. I would not have gotten the job I did without it.

Going in to the bootcamp I was unsure of my decision to enroll. A cursory Google search of coding bootcamps yields a mixed bag when it comes to tech professionals' views on them. A lot of people see bootcamps as a cash grab that churn out low-quality candidates. I was worried about being stigmatized for having a bootcamp on my resume.

For the most part, hiring managers are open to bootcamp graduates. This isn't universally true, but it wasn't a huge hurdle when it came to the job search. The biggest hurdle is the fact that you presumably won't have relevant experience on your resume. In several cases, I felt like I had met the technical bar for a position but was passed over for someone with more experience. This is likely to be a reality in your job search, but it doesn't make getting a job impossible.

That being said, I do believe there is a soft ceiling for jobs you'll be competitive for based on your background. It's not easy to get the Data Scientist title without a graduate degree. Even the more advanced Data Analyst positions seem to be pretty consistently filled by people with an M.S. in a STEM field. The point I’m trying to make is this: if you’re coming in with very little technical experience and no graduate degree, you’re realistically going to be shooting for a Data Analyst role, not a Data Scientist one.

There’s nothing wrong with that though! Getting a Data Analyst role then moving to a Data Scientist role in 2-3 years is a tried and true path in to data science.

Anyways, that’s enough about the reality of the job search. The bootcamp itself was actually a very fun experience. The project-based curriculum is effective in both its delivery of the content and in setting you up to get a job. Being forced to come up with a solution to every problem you encounter is the best way to learn. Prior to the bootcamp I found myself giving up when I encountered something really hard. Having a deliverable and a deadline keeps you a lot more accountable than self-study.

I was expecting to be miserable for the three months of the bootcamp, but it really wasn’t that bad. I’d expect to be putting in 50-60 hours a week, but your days are quite varied. Your mornings are spent doing pair problems and lectures, and your afternoons are spent working. You have 4-5 hours to work on projects every day when you’re on-site – you would be surprised what you can accomplish over a few weeks when you’re throwing that much time in to projects.

The career support was wonderful. Marybeth was super helpful and supportive for all my job needs, but you need to meet her halfway – the career services are not your concierge job lead service.

I’ve come so far in the 6 months since I started the bootcamp. If you enroll, expect lots of personal and professional/technical growth. To be honest, I don’t think I was capable of being as serious as I needed to be about moving in to a more technical job. Metis gave me the structure and help I needed.

One final word of advice: you really do get out of Metis what you put in. This doesn’t mean that you have to go super far out of your comfort zone in terms of the effort you’re putting in, but it does mean you have to be serious about getting in to Data Science. When you encounter problems, you need to be willing to troubleshoot them yourself rather than immediately ask for help. You need to be willing to hold yourself accountable for applying to jobs. The job search can be really tough and downright demoralizing, but if you have persistence you will eventually get where you want to be. Overall, I highly recommend Metis.

Core data science (and I mean rigorous statistical analysis and predictive modeling), felt so out of reach before Metis. There's so many resources for learning Python and SQL out there (Udacity and Mode Analytics being my favs, respectively) - that I wanted material and curriculum from a bootcamp that matched the cost of attendance. In 3 months, Metis made it worth my while - enabling my growth from pivot table repots in excel and simple sql queries all the way to Natural Language Processing on unstructured text and Deep Learning on Cloud Deployments.

The bootcamp option isn't for the faint of heart - and my personal thirst for challenge and rigor was satisfied. The immersive experience was so valuable to me for several reasons:

(a) The "guided tour" experience. There's too many resources for learning data science, and it's helpful to have experienced professionals vet the resources for you. Part of being a good data scientist is knowing what you can down prioritize for learning later. By immersing myself in the best practices and skills that Metis instructors have determined most useful on the job, I generated plenty of leads and had secured an industry role within 3 months of graduating the bootcamp.

(b) Legitimized work experience. It's credible to potential employers to see accredited Metis experience on my resume, with work references from my instructors and cohort members to further validate my experience to potential employers. Whenever a potential employer asked me to list references for someone I worked with, I had a list of capable data scientists that could vouch for me. This set me up for plenty of technical assessments and technical interviews.

(c) Connections to a learning network. Metis alumni are growing with every cohort! Data scientists are a minority compared to Software Engineers, and I was told early on that it wouldn't be unusual to work alone or on small teams on the job. That said, over a year later I'm still leaning on the community to keep my updated as data science tools and methods improve. Post-bootcamp, I learned so much more from my cohort and other alums from soliciting help and resource sharing. If I had done all the same projects by myself, then I wouldn't have had the breadth of exposure or network to be as successful a data scientist as I am.

I can't stress enough how worth it this experience was for me.

I completed my Metis bootcamp in December 2018 in Seattle.  What an amazing 12 week experience.  Every person in the 16 student group was smart, committed and caring as were the Metis staff as well.  We were a diverse group in ages and backgrounds but everyone was there to get the most out of the 12 week experience.  We were together in the classroom from 9am-5pm, 5 days a week.  That made for an amazing energy.  

We learned a lot but it's bootcamp and the information comes at you fast.  This meant we were on a very steep learning curve which is pretty exciting.  Be prepared to be all in for the 12 weeks to get the most out of your time at Metis and then spend time afterwards learning the stuff you missed along the way.

I attended Metis in the summer of 2018 and I could not be happier with my Metis experience. The instruction was state of the art and the staff was accessible, helpful, and supportive.

Probably the most important feature of a data science bootcamp is the quality of the instruction and Metis instructors are excellent. One of my instructors had relatively a stronger background in computer science and the other a relatively stronger background in math and they complemented each other perfectly. The classroom culture was collaborative and supportive. Learning from peers was highly encouraged and practiced daily. Every morning, we completed a pair programming exercise; this was an incredible opportunity to learn from peers who came from diverse backgrounds and had incredibly advanced and varied skillsets. Moreover, updates to the curriculum are continuously being implemented such that the curriculum stays on the cutting edge of data science.

I also found the way the curriculum was structured to be highly effective. The curriculum is project-based and covers topics from exploratory data analysis, to regression to classification to unsupervised learning and finally to a passion project of students’ choosing. Students complete a project for each of these units. Metis does not provide the data sources or provide students with a pre-packaged template for the projects. Students have to go out and scrape their own data or otherwise obtain data from publicly available sources and determine (with guidance from instructors of course) what direction they want to take their projects. Students’ diverse backgrounds and interests are then reflected in the choice of projects; in my cohort alone, student projects ranged from predicting the stock market (of course), to predicting good locations for cat cafes, to natural language processing of religious texts, and predicting locations and severity of wildfires. I found this project-based approach to be highly effective. In contrast to other data science programs which are centered around only one capstone project, Metis’ curriculum allowed for students to take deep dives into each of these topics.

Career support was also extremely effective. From the careers team, I learned how to update my resume and LinkedIn profile to be consistent with industry standards and even practiced in person and technical interviews. I eventually got my first job as a result of a tip from one of the career’s staff. To me, this indicated just how well the career’s staff knows their students; they knew that this opportunity would be a good fit for me and indeed it was.

Of course, Metis is not cheap but none of these bootcamps are. When I made the decision to attend Metis, I knew that I was making an investment in my future and at the time, I predicted (because hey I am a scientist and scientists like to make predictions) that I would have a strong return on that investment. In the few months that it has been since I graduated I have already had a return on that investment and I know that return will continue long into the future.

Lastly, while Metis was definitely an intense and challenging experience, it was also a fun one. It was extremely stimulating to be surrounded everyday by such knowledgeable instructors and peers with such incredibly diverse skillsets. There were late nights and work on the weekends to be sure, but I also found the work to be manageable. During Metis, I remember telling myself not to let perfect be the enemy of good and I think that advice served me well. The skills that I think are critical for students to succeed at Metis are curiosity, a willingness to try something and fail on an initial attempt, a willingness to iterate, and openness to collaboration. If these qualities resonate with you, I would strongly encourage you to check Metis out. It was an amazing experience and I am so glad that I made the choice to do it.

I can't say enough about the curriculum, staff, and community. I completed my course in Fall 2017, I got a job within 2 months. and have continued to benefit from the community as well as the course resources. My background is in psych research in academia. I got my PhD in Psych/Neuroscience and was facing a bleak job market so I made the decision to transition to data science. It was a rough transition. The bootcamp is not for the faint of heart, but my cohort was comprised of really bright and supportive students (many PhDs and graduates with genuine technical and mathematical prowess) who kept me motivated and on-track. My instructors were amazing and the administrators were incredibly hard working, organized and caring. The curriculum was comprehensive (i.e., a lot so be prepared to be overwhelmed). Regardless of how smart/hardworking you are, you will struggle to keep up. BUT if your experience is anything like mine, you'll be surrounded by people who genuinely care about your success. What's impressed me the most is that a year into finding a well-paying tech job, I'm still very much a part of- and benefit from- the Metis community. I STILL use notes from my session. and my cohort still supports each other's growth and success through referrals and knowledge-sharing. The administrators are still available a year plus into graduating for advice and help because they genuinely care about their students. If I could give this place a 6/5 I would. I really loved my time there and I have a ton of respect for the Metis family.

I two took Metis' online courses:

1. Python/Math for Data Science

2. Intro to Data Science

I thought both were very good and definitely worth the money. I was a complete programming beginner but had done some pre-work using Zed Shaw's "Learn Python the Hard Way" and I had taken one intro Python/Data Science class at Simplilearn (very good too).

Class #1 above was the first Metis class I took. Without my preparation I would have been lost, but the prep I did allowed me to keep up with the programming part of the class. The math I fortunately already knew, but it would have been confusing had I not known it. [If you don't know it, then use the class as a menu for what you need to know eventually]. Interaction with the instructor took place via chat window, but he was responsive to Q&A. The course materials (Jupyter Notebooks) were very through and detailed & I've reviewed and re-reviewed afterward and continue to get value from  them.

Class #2 built on Class #1 but got more into techniques of statistical modelling. In this class, we spoke with the instructor (not just chat box), which was very very valuable given the subject matter. I thought the course material (again Jupyter Notebooks) was well structured and clearly showed how to run the models using Python code. The instructor was fantastic, even agreeing to have "office hours" before class and hanging out after class for extra Q&A. Way beyond the call of duty, but it replicated a bit the university environment.

In general, I'd highly recommend either course (and I paid myself, no corporate reimbursement). If -- like me -- you are a total beginner, then do some learning on your own before enrolling. It will be very hard to keep up if you go in cold. They cover a lot of material and go fast. But if you're willing to really try and work hard to learn, I bet you'll find good value and a lot of insight to help you on your path. I did.

 

* Content: It´s an injection of information straight to your brain as expected.  You get to see all of the concepts on Data Science and the stats/math underneath, so be sure to make the prework, and during the bootcamp read more or look for videos on any concept you don't understand very well.  There's a lot of self learning needed: instructors will be there to explain things to you, and you can ask for review sessions on some concepts.  But it's practicing, researching and discussing with your peers that you'll get everything well understood

* Assignments: The most important part are the Projects (every 2/3 weeks) you'll feel you don't have time for anything, but you'll be amazed on how much you learn and get done in a little time, and that's the point.  There are 5 projects: the fist of them given to you, very fast (to deliver in just a week), in group with some of your partners, to get the idea of the bootcamp methodology.  The next 3 will be individually and chosen by you, each of them to practice a particular topic, methodology or algorithm (Regression, Classification, Supervised/Unsupervised learning, NLP, Neural Networks...).  The last one, your "passion project", also individually and chosen by you, can be almost anything you want with what you learn.  So the projects are as hard as you push youlself. 
There are also other assignments: challenges (like homework that will help you with your assignments) every week, a blog and blog posts (and blog posting culture) every now and then, and two personal investigations during the bootcamp... So you´ll think its a lot but it´s all well structured: Make them all, because all assignments will help you in your projects that are the most important part

* Instructors: Since you're learning a wide range of concepts: Statistics, Math, Programming skills, methodologies, libraries... You'll also find different instructors that will give a different focus to the things you are learning.  It can be frustrating sometimes not to get a straight forward answer, but at the end you'll find out that Data Science has a lot of points of view and you have to learn to find the best fit for you, so that's what the instructors are doing by helping you find an answer and not giving it straight away. One thing they'll always help you and support a lot with is the fact that no project is wrong and you can move on with any crazy idea even if you think it's impossible, and they'll show you ways of seeing things that you'll only get by practice and experience.

* Carrer Support: You'll get very useful workshops: to fix your CV, your LinkedIn profile, your networking and interviewing skills, which is great.  There's also people coming over from different companies and Data Science roles to talk about their experience and you can ask them anything!   After bootcamp, people from Carrer Support are still in contact with you for help, advice or any questions until you get a job (Or even after, if you need it)
In my case, that's as far as it goes: I was an international student so other support like: visit to hiring companies, salary negotiation skills or networking with hiring partners was not very useful for me (Although I heard it was very useful for my peers), neither do they offer sponsor for a working visa, that depends on the company and I was not very interested at the time.  However, back home, with what I learned on the workshops, the tips that they gave me and post bootcamp advice I found a job back home in a couple of months.

* Environment: New York's campus is in a WeWork building, there a lot of areas to study in groups or concentrate individually, it's open 24/7 and there are common areas to meet the instructors or your partners regarding bootcamp topics or chat about anything else.

* Conclusion: Bootcamp experience is great. Personally I think is a little expensive, but totally worth it.  You'll get the tools you need to become a Data Scientist and the culture to keep learning and keep growing. If you're an international student, don't expect to get a job to stay in the US and change your life, but you will come back home in only 3 months, having learned a lot of stuff people take maybe a year to learn and succeed.

This review is for intro-to-DS live/online course.

I recently got interested in data science (DS), and I wanted to do a career-transition through the 12-week DS bootcamp offered by METIS. In preparation for the bootcamp, I decided to take the Intro-to-DS live/online course, and I’m glad I did! This course gave me the necessary overview and hands-on coding experience with DS and machine learning (ML) fundamentals. The exploratory analysis techniques I learned in this course helped me tackle the DS challenge given during my (bootcamp) application process.

This Intro-to-DS course is fast-paced and filled with lots of content. The curriculum is structured in such a way that, we could apply the techniques we learn at each session to our project interest and continue to build upon it throughout the course. There was a total of twelve (3-hr) live sessions where we got to learn basic concepts of DS, i.e., math (LinAlg & Stats) and python, supervised- & unsupervised learning, and pretty much everything involved in a typical model-building pipeline. By the way, twelve sessions surely won’t be enough to cover all existing ML techniques today, but at least we got to see a high-level overview of most of them. Each session was held live and online via ZOOM (where the instructor and students could share each other’s screen); these sessions were also recorded so we were able to download the videos after. Course materials (e.g., lecture notes, jupyter notebooks) were distributed via git-hub, so I could keep these notes for future reference. I wished there were more homework sets given to help solidify the concepts we learned in class, but I suppose I could just google search for various datasets and explore them myself.

During the course, we were strongly encouraged to complete a mini DS project. About half the class (including myself) selected a project, worked on it throughout the course and presented it at our last session. This was a good experience for me, as I had never worked on an end-to-end DS project before. I’ve taken a couple of self-paced MOOC’s, but the live (online) classroom setting in this course was more helpful for me, as I somehow felt accountable to keep up with the materials as the course progresses. 

I’d recommend taking this course, if you have some programming experience with python and are interested in getting hands-on coding experience with ML. By the way, the tuition for this course can be applied to the bootcamp, if accepted to the program. So, it’s basically a free, pre-bootcamp (warm-up) training.