Elisabeth Günther

The Road to MLOps: How Verivox Transitioned to AWS

Plagued by months-long deployment cycles, Verivox transformed their ML workflow. See how they used AWS CDK and MLOps to achieve deployments in just hours.

The Road to MLOps: How Verivox Transitioned to AWS
#1about 3 minutes

Understanding the role and challenges of MLOps

MLOps provides a structured process to build and integrate machine learning products by addressing challenges beyond just the ML code, such as versioning, security, and deployment.

#2about 4 minutes

Navigating the four phases of MLOps maturity

The MLOps maturity model guides teams through four phases: accelerating proof of concept, making processes repeatable, ensuring reliability through monitoring, and achieving scalability.

#3about 3 minutes

Overcoming siloed code and deployment bottlenecks

Verivox's initial setup suffered from siloed codebases, a lack of deployment ownership, and friction between teams, prompting a complete operational transformation.

#4about 2 minutes

Executing a multi-stage initial migration to AWS

The team's first project involved migrating from R to Python and moving from manual UI clicks to a fully automated CI/CD pipeline with infrastructure as code.

#5about 3 minutes

Building a real-time inference architecture on AWS

A standardized blueprint using Amazon SageMaker Pipelines and AWS Lambda was created to solve the major pain point of deploying models for real-time inference.

#6about 2 minutes

Using AWS Fargate for flexible batch processing

A container-based architecture with AWS Fargate and Step Functions provides the flexibility needed for custom batch jobs and lifting-and-shifting legacy projects.

#7about 4 minutes

Automating infrastructure with AWS CDK templates

AWS Cloud Development Kit (CDK) enables the creation of reusable, parameterizable infrastructure templates to scale deployments across multiple projects, accounts, and sandboxes.

#8about 3 minutes

Key learnings and results from the MLOps transformation

The migration resulted in drastically reduced deployment times, improved reliability, and new capabilities, underscoring the value of support networks and managed services.

Related jobs
Jobs that call for the skills explored in this talk.

Matching moments

Creating a stepwise transition strategy to MLOps
03:14 MIN

Creating a stepwise transition strategy to MLOps

Effective Machine Learning - Managing Complexity with MLOps

The convergence of ML and DevOps in MLOps
01:33 MIN

The convergence of ML and DevOps in MLOps

AI Model Management Life Circles: ML Ops For Generative AI Models From Research to Deployment

Choosing between a custom vs managed MLOps platform
09:14 MIN

Choosing between a custom vs managed MLOps platform

Effective Machine Learning - Managing Complexity with MLOps

Q&A on migration strategy and stakeholder management
10:39 MIN

Q&A on migration strategy and stakeholder management

AWS Migration within 3 months

Q&A: MLOps tools for building CI/CD pipelines
02:49 MIN

Q&A: MLOps tools for building CI/CD pipelines

Data Science in Retail

Implementing the end-to-end MLOps pipeline
01:45 MIN

Implementing the end-to-end MLOps pipeline

How We Built a Machine Learning-Based Recommendation System (And Survived to Tell the Tale)

Using data management and open source tools for MLOps
02:18 MIN

Using data management and open source tools for MLOps

MLOps - What’s the deal behind it?

Automating the data pipeline with multi-cloud services
04:09 MIN

Automating the data pipeline with multi-cloud services

Leverage Cloud Computing Benefits with Serverless Multi-Cloud ML

Featured Partners

Related Videos

See all videos
Building the platform for providing ML predictions based on real-time player activity
30:57

Building the platform for providing ML predictions based on real-time player activity

Artem Volk & Fabian Zillgens

AWS Migration within 3 months
37:08

AWS Migration within 3 months

Steffen Heilmann

Remote Driving on Plant Grounds with State-of-the-Art Cloud Technologies
48:54

Remote Driving on Plant Grounds with State-of-the-Art Cloud Technologies

Oliver Zimmert

Leverage Cloud Computing Benefits with Serverless Multi-Cloud ML
39:58

Leverage Cloud Computing Benefits with Serverless Multi-Cloud ML

Linda Mohamed

We adopted DevOps and are Cloud-native, Now What?
34:46

We adopted DevOps and are Cloud-native, Now What?

Bruno Amaro Almeida

DevOps for Machine Learning
44:40

DevOps for Machine Learning

Hauke Brammer

Effective Machine Learning - Managing Complexity with MLOps
45:45

Effective Machine Learning - Managing Complexity with MLOps

Simon Stiebellehner

DevOps for AI: running LLMs in production with Kubernetes and KubeFlow
34:21

DevOps for AI: running LLMs in production with Kubernetes and KubeFlow

Aarno Aukia

Related Articles

View all articles
Benedikt Bischof
MLOps – What’s the deal behind it?
Welcome to this issue of the WeAreDevelopers Live Talk series. This article recaps an interesting talk by Nico Axtmann who introduced us to MLOpsAbout the speaker:Nico Axtmann is a seasoned machine learning veteran. Starting back in 2014 he observed ...
MLOps – What’s the deal behind it?
Benedikt Bischof
MLOps And AI Driven Development
Welcome to this issue of the WeAreDevelopers Dev Talk Recap series. This article recaps an interesting talk by Natalie Pistunovic who spoke about the development of AI and MLOps. What you will learn:How the concept of AI became an academic field and ...
MLOps And AI Driven Development

From learning to earning

Jobs that call for the skills explored in this talk.

Python Engineer

Python Engineer

Usersnap
Vienna, Austria

Intermediate
Python
Amazon Web Services (AWS)