Operationalizing Machine Learning at Scale

Operationalizing Machine Learning at Scale

Aligning DevOps and AI/ML development with scalable and
secure end-to-end data and model pipeline automation

COURSE : Operationalizing Machine Learning at Scale
DATE/S : 6th/7th of November 2019
VENUE: HYPERIGHT DATA CLUB | STOCKHOLM Tegnergatan 14, 113 58 Stockholm


It is hard to build production systems that rely on bleeding-edge technology. The technology stacks frequently change and the goal is often not a static product but rather one that adapts rapidly to accommodate new tools and data.

The adoption of DevOps has been very effective at bringing together software development and operations teams to simplify and improve deployment and release processes. As Artificial Intelligence (AI) and machine learning (ML) become increasingly more important components for applications, more pressure will exist to ensure they are part of an organization’s DevOps model.

The challenge is getting the ML model deployed into a production environment and keeping it operational and supportable. Software development teams know how to deliver business applications and cloud services. AI/ML teams know how to develop models that can transform a business. But when it comes to putting the two together to implement an application pipeline specific to AI/ML — to automate it and wrap it around good deployment practices — the process needs some effort to be successful. This integration process requires ML pipelines to be compatible with the state of the art platforms that can run production grade services with strong service level agreements with essential features like scalability, security, efficiency and fault tolerance.

Scaleout Lean AI let’s data science become part of a continuous delivery system together with domain experts, software developers and IT operations and introduces an efficient and sustainable method for getting machine learning models deployed into a production environment and keeping them operational and supportable. The aim of Lean AI is to establish processes and technology that integrate knowledge, processes and data in workflows supporting continuous innovation. On a technical level, Lean AI promotes the use of best-of-breed open source cloud native technology for vendor agnostic solutions and for long term sustainability.


Why this training is crucial for your organisation:

This training aims to provide attendees with practical knowledge of how to operationalize machine learning at scale. Scaleout Lean AI is designed to accelerate the adoption of AI by empowering the teams that take use cases to production services. It addresses the challenge of putting and keeping AI in production at its core by providing teams with knowledge, tools and support needed to repeatedly succeed with AI projects and ultimately delivering business value quickly and efficiently.

This two day intensive and interactive training will bring to you a unique outlook on aligning DevOps and AI/ML development. The course boasts both highly experienced software product developers with many years of ex- perience delivering solutions across industries, as well as highly acclaimed instructors from academia with many years applied experience in data science, cloud infrastructure, machine learning and AI. The intrinsic knowledge and practical coaching methods will equip you with skills that you will be able to implement immediately in your organisation. Do not miss this opportunity of learning powerful methods for taking AI to production.

Benefits of attending this interactive two-day training include

Aligning your AI strategy with the objectives of the business
Defining the roles of the various departments and streamlining the roles for effective AI-management
Managing changing requirements for streamlined management
Encompassing KPI’s into your ML model process
Determining whether the deliverables of the ML process can be linked to your KPI’S
Measuring ML model performance
Dealing effectively with non-performance
Benchmarking services and establishing world class standards
Defining flexible and manageable communication interfaces

*The workshop will cover essential high-level details required to design and understand systems architecture based on state of the art open source technologies. This workshop will not include in-depth technical details required to implement or deploy machine learning pipelines.

Who should attend

Line managers, digitalisation leaders, data and analytics managers, Tech Leads, IT Project Heads. Current or future participants in machine learning projects – domain experts, software developers, data scientists and IT operations. From all industry sectors.

The Scaleout training programme methodology

Our courses are thoroughly researched and structured to provide intense and intimate practical training to your organisation. Our format consists in:

  • Pre-course questionnaires
  • An in-depth tailored programme to address market concerns Diverse real-life case examples
  • Comprehensive course documentation
  • Interactive roundtable discussion and breakout sessions Hands-on “learning by doing”

Training needs analysis & post training effectiveness assessment

To ensure that participants gain maximum benefits from the training, detailed questionnaires will be sent to all course participants to establish exactly where their training needs lie. The completed forms will be analysed by the course trainer. As a result, we ensure deliverance at the appropriate level and issues participants regard as relevant are addressed. The comprehensive course materials will enable them to digest the subject matter in their own time. The second Questionnaire would be a Post Training Effectiveness Assessment. The purpose of this questionnaire would be to measure and assess the learning outcomes with the participants a month after the training to gauge their level of understanding in implementing what they have learnt. Thus, identifying further areas for improvement.


Session One
– Overview of the stages of ML development
– Machine learning problem framing and solution proposal
– Model Serving and management life cycle Importance of selecting the right technologies Storage solutions
– Management of APIs
– Security modules
– Data Preparation and Feature Engineering in ML Role of DevOps in Model management and serving pipelines
– Testing and Debugging in Machine Learning

Session Two
Interactive Discussion:
– Discussion around the topics covered in session one
– Participants will introduce themselves and talk about their AI related challenges

Session Three
Required technical details to design a scalable architecture
– Choice of resources
– In-house and public clouds
– Virtualization and Containers
– Bare metal
– Choice of platform
– Keras
– Scikit-learn
– Theano
– Caffe
– Torch
– Choice of services
– Provider-specific and open-source services
– Requirements of a production grade environment
– Reliability

Session Four
In this session, participants will benefit from a question and answer exercises on all topics discussed during the first day. It will be interactive, allowing for analysis of questions and answers.

Session Five
Introduction to different case studies

Session Six
This session will be based on a group activity where each group consists of two to three participants will architect a solution for one of the case studies presented in session four. The desired solution needs to be based on the following three components:
– Identify the requirements
– Choose adequate infrastructure, platform and services to design an end to end production grade ML service
– Based on the discussed technologies, design the architecture diagram

Each group is supposed to present the solution and explain the design choices. Together with instructors, the rest of the participants will discuss and give their feedback.

Instructors & co-instructors

Morgan Ekmefjord
Position : CTO
Company: Scaleout Systems

Business and Product development experience from several Fortune 500 companies. Successfully built solutions ranging across a wide range of industries from Mission planning and Fighter simulators, Telecommunication planning and optimization software, Regulatory and Compliance tracking systems, Large Scale POS solutions for retail and food companies.
Ola Spjuth

Position : Lead Scientist ML and AI
Company : Scaleout System

Ola is the lead scientist machine learning & AI at Scaleout Systems and holds a PhD in bioinformatics with many years of experience
in applied machine learning on high-performance and distributed e-infrastructures. Ola is also Associate Professor at Uppsala Univer- sity where he leads a research group that studies how predictive modelling, large-scale calculations and modern e-infrastructure can aid research and development.
Salman Toor
Position : Lead Scientist, Distributed Infrastructures Company : Scaleout Systems

He holds a PhD in scientific computing and is an expert on scientific data management, scalability and performance of distributed infrastructures, and solutions for data-intensive applications. Salman is also Assistant professor at Uppsala University where he conducts research on e-infrastructure.


As a research and innovation partner we help you turn your visions into the intelligent applications of tomorrow. We are experienced cloud architects, DevOps engineers and data scientists that can help ensure that your project runs smoothly and that the developed solutions are designed in a sustainable way. 



COURSE : Operationalizing Machine Learning at Scale
DATE/S : 6th/7th of November 2019
Tegnergatan 14, 113 58 Stockholm


Price Per Delegate : 27 390 SEK (excl. VAT) Closing
Registration Date: 25th October 2019

All options inclusive of course papers, luncheon, refreshments & service charge.

After receiving payment a receipt will be issued. If you do not receive a letter outlining joining details two weeks prior to the event, please contact the training coordinator at Dairdux.

    Name of participant/s

    Invoicing Details


    1. Fees are inclusive of programme materials and refreshments.

    2.Payment Terms: Following completion and return of the registration form, full payment is required within 30 days from receipt of invoice or one day before the start of the training. A receipt will be issued on payment. Due to limited training space, we advise early registration to avoid disappointment. A 50% cancellation fee will be charged under the terms outlined below. We reserve the right to refuse admission if payment is not received on time.

    3.Cancellation/Substitution: Provided the total fee has been paid, substitutions at no extra charge up to 7 days before the event are allowed. Substitutions between 7 days and the date of the event will be allowed subject to an administration fee of equal to 10% of the total fee that is to be transferred. Otherwise all bookings carry a 50% cancellation liability immediately after a signed sales contract has been received by Dairdux (as defined above). Cancellations must be received in writing by mail six (6) weeks before the conference is to be held in order to obtain a full credit for any future Dairdux trainings. Thereafter, the full training fee is payable and is nonrefundable. The service charge is completely non-refundable and non-creditable.Payment terms are thirty days and payment must be made prior to the start of the conference. Non-payment or non-attendance does not constitute cancellation. By signing this contract, the client agrees that in case of dispute or cancellation of this contract that Dairdux will not be able to mitigate its losses for any less than 50% of the total contract value. If, for any reason, Dairdux decides to cancel this training, full fee will be refunded. Dairdux is not responsible for covering airfare, hotel, or other travel costs incurred by clients. Event programme content is subject to change without notice.

    4.Copyright etc: All intellectual property rights in all materials produced or distributed by Dairdux inconnection with this event is expressly reserved and any unauthorized duplication, publication or distribution is prohibited.

    5.Client information is kept on Dairdux group companies database and used by Dairdux group of companies to assist in providing selected products and services which maybe of interest to the Client and which will be communicated by phone or email. If you do not want Dairdux to do this please tick this box [ ]. For training and security purposes telephone calls maybe recorded.

    6.Important note: While every reasonable effort will be made to adhere to the advertised package, Dairdux reserves the right to change event dates, sites or location or omit event features, or merge the event with another event, as it deems necessary without penalty and in such situations no refunds, part refunds or alternative offers shall be made. In the event that Dairdux permanently cancels the event for any reason whatsoever, (including, but not limited to any force majeure occurrence) and provided that the event is not postponed to a later date nor is merged with another training, the Client shall receive a full refund of the amount that the Client has paid to such permanently cancelled event, or receive a credit note for the amount payed for up to one year to be used at another Dairdux event.

    7.Governing law: This Agreement shall be governed and construed in accordance with the law of Sweden and the parties submit to the exclusive jurisdiction of the Swedish Courts in Sweden.

    8. Client hereby acknowledges that that this Contract is valid, binding and enforceable; and that he/she has no basis to claim that any payments required under this Contract at any time are improper, disputed or unauthorized in any way. Client acknowledges that they have read and understood all terms of this contract, including, without limitation, the provisions relating to cancellation.

    Catch a cookie!

    This website uses cookies to help improve your user experience.