We don't just train models, we solve complex and business-critical ML problems.
TRAINING FORMAT
ONLINE
WHO'S FITTING
MIDDLE/SENIOR
PACKAGE OFFERS
1 TO 5 BLOCKS
WHO THIS PROGRAMME IS FOR:
ML DEVELOPER
Have experience in machine learning related fields and want to learn how to solve specific and non-standard problems.
TIMLID
Manage an ML development team and want to learn best practices for building services to solve complex business problems.
RECOMMENDED LEVEL:
THEORY
INFRA-STRUCTURE
> Knowledge of the basics of probability theory and mathematical statistics
> Understanding of A/B-tests and Bootstrap method
> Knowledge of classical ML algorithms
> Basic knowledge of Bash, Docker, Git
> Minimum experience with web frameworks (e.g. Flask)
PYTHON
> Knowledge of language syntax
> Understanding of basic data structures (list, dictionary, tuple)
> Knowledge of OOP basics (class, object)
> Knowledge of Pandas, Scikit-learn, and optionally PyTorch libraries.
> Ability to debug code and write tests
COURSE PROGRAMME ://
Algorithms for automatic matching of objects allow solving many important business tasks: identifying duplicate products, combining online and offline customer data, analysing and monitoring competitors' prices. We will look at various cases of applying matching and ranking systems, consider the technical subtleties of their architectures, get acquainted with advanced approaches to model training and implement our own system of search hints.
10 lessons and a final project
The efficiency of many companies depends on the ability to set optimal prices for products, taking into account various factors and changes in the market. We will learn how to predict price ranges, review metrics relevant to pricing, get to know the "multi-armed bandits" and train one of them to choose a pricing strategy to optimise revenue and sales margins.
9 lessons and a final project
Consumer segmentation by the net effect of marketing influence allows a business to focus its advertising budget on customers who are ready to perform a target action only in the presence of communication. We will learn how to estimate the value of this effect using uplift models, analyse interesting cases from practice, design a library of factors and build our own algorithm for finding the optimal audience.
5 lessons and a final project
Important tasks in A/B testing are speeding up tests and increasing their sensitivity. This allows companies to implement positive changes faster and abandon negative ones in time. We will learn how to apply modern methods of increasing the sensitivity of A/B-tests, consider a complete testing pipeline and implement our own service for evaluating experiments.
10 lessons and final project
It is important for ML-engineers to be able to integrate their solutions into the existing infrastructure, taking into account the high performance and fault tolerance requirements of applications. We will consider different scenarios of service deployment, discuss the principles of their design and maintenance from the perspective of a modern backend, and learn how to properly bring applications into production.
9 lessons and final project
HOW TRAINING TAKES PLACE
SOLVE REAL PROBLEMS
- Solve complex and non-standard ML problems - Learn models and customise everything you need to make them work - Create your own ML services and learn how to bring them into production
UTILISE OUR INFRASTRUCTURE
- Work on a dedicated server - Practise on data from real problems - Learn solutions from instructors and other students
ASK ANY SUPPORT QUESTIONS
- Discuss tasks and projects with market experts - Your mentors will be ML-engineers from leading Russian companies
GET A JOB AFTER THE COURSE
- Consult with HR specialists - Learn how to properly present your skills - Apply for vacancies of partner companies - Go on interviews and get a job
STUDENT FEEDBACK/
High quality material and assignments
Extremely useful courses on advanced data analysis. I would especially like to highlight the modules on A/B testing and Uplift modelling. The quality of the material and the tasks are at a high level.
This course had a global impact on my career. Thanks to the acquired skills, especially the module on advanced A/B testing, 2 months after the end of the course I received an offer for the position of Senior research analyst in the largest marketplace in Russia, although I had not worked in IT companies before.
Clifford Hansen
Useful course with a practical focus
I liked the practical orientation. There were enough modern approaches, theory and tasks in the selected topics.
I liked the fact that there were analyses of assignments - it really helped. I also liked the interactivity with lecturers and other students. Very useful course, keep it up!
Some things I was able to apply immediately, some I watched diagonally. In general, I don't use all the tasks of the course now, but some sections (deployment, pricing) I have already managed to apply in my work. Some things I just haven't had time to understand properly yet (the same A/B tests).
Bill Jones
The knowledge came in handy both in interviews and in the job itself
I took advantage of the employment assistance programme included in the course. And it was through the specialist karpov. courses my new job found me (although I was looking for vacancies myself). I got a job in a small, cosy but promising start-up, where I developed from scratch a credit scoring product for small entrepreneurs to sell to banks.
Before I started looking for a job, I only passed the modules "Ranking and Matching" and "Dynamic Pricing". And this knowledge came in handy for me both in interviews and in my work. Also, of course, the "ML-Service Deployment Scenarios" help me in my daily work.
I haven't had a chance to apply the title topics of the modules "Uplift-modelling" and "Advanced A/B-testing" directly in practice yet, but the acquired knowledge was not superfluous anyway: advanced A/B-testing improved statistics, and in the module on uplift there was a topic on designing a library of factors, which is actual in any ML-tasks.
In general, each module had something new, useful, interesting for me. But if we talk about personal preferences, I especially liked "Ranking and Matching" and "Uplift modelling". The charm of the teachers (no offence to the teachers of other blocks), the structure of the modules, and some of my own interests played a role here.
John Johnson
Enjoyed solving complex problems
Interesting material: you can learn a lot of new things from the course that can only be found in specialised literature. Difficult tasks: I had to sit on many tasks for many hours. But I was really happy when I finally managed to reach the solution on my own.
I joined the course because I wanted to broaden my horizons by solving non-typical problems. Unfortunately, there is no practical application of the acquired skills at my current place of work. I would especially like to thank Vladislav Ladenkov for not only recording excellent lectures, but also personally discussing the final project with me in detail.
Earl Boland
TUITION FEE
Immerse yourself in the world of advanced ML, access to remote server work and support from our instructors.
> Ranking and Matching > Dynamic pricing > Uplift modelling
> Advanced A/B-testing > ML Services Deployment Scenarios
To pay for the course, you need to register on our education platform with your first name, last name and email.
If you already have an account, you can use it.
ASK A QUESTION
We will contact you and answer any questions you may have about the course.
FAQ
Yes, we carry out educational activities on the basis of a state licence.
For comfortable learning on the course you need to be able to write code in Python and compose SQL queries to databases. You will not need specialised knowledge of data engineering.
Yes, you can now buy only the blocks you are currently interested in. The more blocks you buy, the more favourable the cost of each block will be. You can read more about the cost here. Later you will be able to buy more blocks at a special price of $450.
For each block, we give you 1.5 months of access to support and infrastructure. You can distribute this time among the blocks, for example, you can take them in parallel. After this time, access to support and infrastructure will be cancelled, but the course materials will stay with you forever.
Yes, if you have bought a package of several blocks, you will get access to the first lessons of all paid blocks on the start day. You choose which block you want to start with, and you can take the blocks in parallel.
On average, our students study 10-15 hours per week. This is enough time to be able to watch lectures and complete homework on time. However, the workload of all modules is different and depends on your level of preparation - if you are already familiar with the models and algorithms underlying the module, it will take less time to complete it.
The training takes place on our own LMS educational platform. You will be able to watch pre-recorded lectures, take notes, complete assignments on our infrastructure, and watch lessons. Lessons will open sequentially based on your progress. At the end of each module you will have a final project that will help you consolidate what you have learnt and will be a good addition to your CV. If there are any difficulties during the training, you can ask for help from the guys in the support team. We also hold Q&A sessions with Valery Babushkin for our students every month.
You can watch the lectures from any device, but you will need a computer or laptop to write code. There are no requirements for configuration and power - we will provide all the necessary infrastructure to work on a remote server. At the start of training, you don't need to install any special software - you will only need a browser and standard applications for communication: Discord and Zoom.
We have organised the training in such a way that you can combine it with your work, study and personal life. You can study at any time and at a pace that suits you.
It is quite normal to get "stuck" on a task during training. In this case, we have a support team that will help you to solve a difficult task.
If things do not go according to plan and you feel that you are falling behind on the programme, please let the course supervisors know. Together we will find ways to make your learning experience more convenient.
ANY QUESTIONS?
Fill out the form, we will contact you, answer all your questions and tell you more about the course.