SlideShare a Scribd company logo
1 of 24
DevOps for AI Apps
Richin Jain, Software Engineer (@richinjain)
Vivek Gupta, Data Scientist (@gkeviv)
Agenda
• Background
• DevOps Introduction
• Enterprise AI use case
• Workflows
• Traditional v/s AI App
• Proposed approach
• Pipeline
Background
• Over the course of engaging with various enterprise customers on their AI use case, this has been a
common ask.
• At one point in time Data Science was one off task where answer was given.
• Now it has been integrated with real time applications along with retraining and A/B testing.
• Because of this we have to relook at Data Science process and how it could be integrated with existing
software stack.
• Goal was to start looking at best practices from S/W engineering and how could it best applied here.
DevOps brings together people, processes, and technology, automating software delivery to provide continuous value to
your users.
What is DevOps?
Continuous Integration (CI)
• Focuses on blending the work of
individual developers together into a
repository.
• Each time you commit code, it’s
automatically built and tested and
bugs are detected faster.
Continuous Deployment (CD)
• Automate the entire process from
code commit to production if your
CI/CD tests are successful.
Continuous Learning & Monitoring
• Using CI/CD practices, paired with
monitoring tools, safely deliver features
to your customers as soon as they’re
ready.
Single environment
Multiple environment
Configure Code Build Test Package Deploy
Monitor
• Infrastructure as Code
• Continuous Integration
• Automated Testing
• Continuous Deployment
• Release Management
• Load Testing & Auto-Scale
• App Performance Monitoring
DevOps Maturity
Data Science
• Experimentation
• Modeling
• Versioning
• Lineage
• Conversion
• Export
• Quantization
• Inferencing
• Retraining
• A/B Testing
A C T I V I T I E S
• Need to solve ML problem quickly.
• ML stack might be different from rest of the
application stack.
• Lots of glue code.
• Testing accuracy of ML model.
• ML code is not always version controlled.
• Hard to reproduce models
• Integrating model into application can take weeks
• Need to re-write featurizing and scoring code
multiple times (in different languages)
• Want to start using customer data to build models
• Hard to track breaking changes
P A I N P O I N T S
Enterprise AI use case
• Contoso LLC has an image recognition scenario. Data Science team develops a
state-of-the-art image recognition model.
• Four ways it could be consumed
• User upload the images to Contoso's website and get instant results.
• User uploads several images or point to a folder and get results.
• Native mobile app
• Edge devices
API Based model integration
• Real time
• Batch
Embedded models
• Native Apps
• Edge devices
Workflow: App DeveloperWorkflow: App DeveloperBasic Workflow: Software Engineer
Workflow: Data ScientistWorkflow: App DeveloperBasic Workflow: Data Scientist
https://ai.google/research/pubs/pub45742
https://ai.google/research/pubs/pub45742
Model TestsData Tests
ML Infrastructure Tests Monitoring Tests
Proposed Approach
Data-
Science-
Repo
Publish
test
results
Get
Source
Code
Install
Requirements
Create Conda
Environment
Unit-testPylint
Data
Scientist
Code
Coverage
Data-
Science-
Repo
Get
Source
Code
Install
Requirements
Create Conda
Environment
Unit-test
Create
Docker
Image
Data
Scientist
Register
Model
Pull Req. Pass
Test
deployed
image
Deploy on
Test
Model
Testing and
Validation
Build
Artifact
Create Conda
Environment
Install
Requirements
Deploy to Test
(create/update)
Test environment
(continuous
deployment)
Dev
Artifact
Create Conda
Environment
Install
Requirements
Deploy to
Staging
(update)
Staging Environment
(nightly, other
services test here)
Test
Artifact
Create Conda
Environment
Install
Requirements
Deploy to Prod
(update)
Prod environment
(end of sprint)
Get
Source
Create Conda
Environment
Install
Requirements
Convert model
to other formats
ONNX, CoreML, WinML
(end of sprint, every time
there is a new model)
Get
Source
Create Conda
Environment
Install
Requirements
Retrain model
on new data
Retraining Pipeline
(every night, or triggered
on new data uploading to
blob)
It will run in a pre-prod
environment, so it has
access to production
data, and wouldn’t be
promoted unless it passes
A/B tests against prod
data.
Data
Validation
A/B Testing
Model
Testing and
Validation
Model
Management &
Promotion
• Core features of Azure ML service
exposed through a Python SDK and CLI.
• Easy and simple pip install
• Makes CI/CD much simpler.
• Best practices for architecting and managing an enterprise-ready AI application
lifecycle.
• Azure DevOps and Azure ML ease the adoption of DevOps by DS teams.
• Adoption will increase the agility, quality and delivery of DS teams.
Thank you !

More Related Content

What's hot

What's hot (20)

AWS Summit - Trends in Advanced Monitoring for AWS environments
AWS Summit - Trends in Advanced Monitoring for AWS environmentsAWS Summit - Trends in Advanced Monitoring for AWS environments
AWS Summit - Trends in Advanced Monitoring for AWS environments
 
Boston DevOps Days 2016: Implementing Metrics Driven DevOps - Why and How
Boston DevOps Days 2016: Implementing Metrics Driven DevOps - Why and HowBoston DevOps Days 2016: Implementing Metrics Driven DevOps - Why and How
Boston DevOps Days 2016: Implementing Metrics Driven DevOps - Why and How
 
Applying AI to Performance Engineering: Shift-Left, Shift-Right, Self-Healing
Applying AI to Performance Engineering: Shift-Left, Shift-Right, Self-HealingApplying AI to Performance Engineering: Shift-Left, Shift-Right, Self-Healing
Applying AI to Performance Engineering: Shift-Left, Shift-Right, Self-Healing
 
Metrics-driven Continuous Delivery
Metrics-driven Continuous DeliveryMetrics-driven Continuous Delivery
Metrics-driven Continuous Delivery
 
DevOps: Cultural and Tooling Tips Around the World
DevOps: Cultural and Tooling Tips Around the WorldDevOps: Cultural and Tooling Tips Around the World
DevOps: Cultural and Tooling Tips Around the World
 
Release Readiness Validation with Keptn for Austrian Online Banking Software
Release Readiness Validation with Keptn for Austrian Online Banking SoftwareRelease Readiness Validation with Keptn for Austrian Online Banking Software
Release Readiness Validation with Keptn for Austrian Online Banking Software
 
OOP 2016 - Building Software That Eats The World
OOP 2016 - Building Software That Eats The WorldOOP 2016 - Building Software That Eats The World
OOP 2016 - Building Software That Eats The World
 
Top Java Performance Problems and Metrics To Check in Your Pipeline
Top Java Performance Problems and Metrics To Check in Your PipelineTop Java Performance Problems and Metrics To Check in Your Pipeline
Top Java Performance Problems and Metrics To Check in Your Pipeline
 
Taking the Best of Agile, DevOps and CI/CD into security
Taking the Best of Agile, DevOps and CI/CD into securityTaking the Best of Agile, DevOps and CI/CD into security
Taking the Best of Agile, DevOps and CI/CD into security
 
Monitoring as a Self-Service in Atlassian DevOps Toolchain
Monitoring as a Self-Service in Atlassian DevOps ToolchainMonitoring as a Self-Service in Atlassian DevOps Toolchain
Monitoring as a Self-Service in Atlassian DevOps Toolchain
 
Continuous delivery in Qbon
Continuous delivery  in QbonContinuous delivery  in Qbon
Continuous delivery in Qbon
 
Four Practices to Fix Your Top .NET Performance Problems
Four Practices to Fix Your Top .NET Performance ProblemsFour Practices to Fix Your Top .NET Performance Problems
Four Practices to Fix Your Top .NET Performance Problems
 
DOES SFO 2016 - Scott Willson - Top 10 Ways to Fail at DevOps
DOES SFO 2016 - Scott Willson - Top 10 Ways to Fail at DevOpsDOES SFO 2016 - Scott Willson - Top 10 Ways to Fail at DevOps
DOES SFO 2016 - Scott Willson - Top 10 Ways to Fail at DevOps
 
Top Performance Problems in Distributed Architectures
Top Performance Problems in Distributed ArchitecturesTop Performance Problems in Distributed Architectures
Top Performance Problems in Distributed Architectures
 
Keptn - Automated Operations & Continuous Delivery for k8s
Keptn - Automated Operations & Continuous Delivery for k8sKeptn - Automated Operations & Continuous Delivery for k8s
Keptn - Automated Operations & Continuous Delivery for k8s
 
Quality Jam 2017: Elise Carmichael and Corey Pyle "Jumpstarting Your Test Aut...
Quality Jam 2017: Elise Carmichael and Corey Pyle "Jumpstarting Your Test Aut...Quality Jam 2017: Elise Carmichael and Corey Pyle "Jumpstarting Your Test Aut...
Quality Jam 2017: Elise Carmichael and Corey Pyle "Jumpstarting Your Test Aut...
 
Quality Jam 2017: Kevin Dunne "Macro Trends and Useful Tools that 'Get It'"
Quality Jam 2017: Kevin Dunne "Macro Trends and Useful Tools that 'Get It'"Quality Jam 2017: Kevin Dunne "Macro Trends and Useful Tools that 'Get It'"
Quality Jam 2017: Kevin Dunne "Macro Trends and Useful Tools that 'Get It'"
 
Deploy Faster Without Failing Faster - Metrics-Driven - Dynatrace User Groups...
Deploy Faster Without Failing Faster - Metrics-Driven - Dynatrace User Groups...Deploy Faster Without Failing Faster - Metrics-Driven - Dynatrace User Groups...
Deploy Faster Without Failing Faster - Metrics-Driven - Dynatrace User Groups...
 
Web and App Performance: Top Problems to avoid to keep you out of the News
Web and App Performance: Top Problems to avoid to keep you out of the NewsWeb and App Performance: Top Problems to avoid to keep you out of the News
Web and App Performance: Top Problems to avoid to keep you out of the News
 
Managers, Future Proof Your Automation
Managers, Future Proof Your AutomationManagers, Future Proof Your Automation
Managers, Future Proof Your Automation
 

Similar to DevOps for AI Apps

SanDiego_DevOps_Meetup_9212016
SanDiego_DevOps_Meetup_9212016SanDiego_DevOps_Meetup_9212016
SanDiego_DevOps_Meetup_9212016
w2fong
 
SanDiego_DevOps_Meetup_9212016-v8
SanDiego_DevOps_Meetup_9212016-v8SanDiego_DevOps_Meetup_9212016-v8
SanDiego_DevOps_Meetup_9212016-v8
Rajwinder Singh
 

Similar to DevOps for AI Apps (20)

Application Lifecycle Management
Application Lifecycle ManagementApplication Lifecycle Management
Application Lifecycle Management
 
DevOps for Machine Learning overview en-us
DevOps for Machine Learning overview en-usDevOps for Machine Learning overview en-us
DevOps for Machine Learning overview en-us
 
DevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer Tools
DevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer ToolsDevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer Tools
DevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer Tools
 
DevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer Tools
DevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer ToolsDevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer Tools
DevOps on AWS: Deep Dive on Continuous Delivery and the AWS Developer Tools
 
DevOps on AWS - Building Systems to Deliver Faster
DevOps on AWS - Building Systems to Deliver FasterDevOps on AWS - Building Systems to Deliver Faster
DevOps on AWS - Building Systems to Deliver Faster
 
SRV312 DevOps on AWS: Building Systems to Deliver Faster
SRV312 DevOps on AWS: Building Systems to Deliver FasterSRV312 DevOps on AWS: Building Systems to Deliver Faster
SRV312 DevOps on AWS: Building Systems to Deliver Faster
 
Dev ops using Jenkins
Dev ops using JenkinsDev ops using Jenkins
Dev ops using Jenkins
 
A Tale of Two Pizzas: Accelerating Software Delivery with AWS Developer Tools
A Tale of Two Pizzas: Accelerating Software Delivery with AWS Developer ToolsA Tale of Two Pizzas: Accelerating Software Delivery with AWS Developer Tools
A Tale of Two Pizzas: Accelerating Software Delivery with AWS Developer Tools
 
SanDiego_DevOps_Meetup_9212016
SanDiego_DevOps_Meetup_9212016SanDiego_DevOps_Meetup_9212016
SanDiego_DevOps_Meetup_9212016
 
DevOps on AWS: DevOps Day San Francisco
DevOps on AWS: DevOps Day San FranciscoDevOps on AWS: DevOps Day San Francisco
DevOps on AWS: DevOps Day San Francisco
 
Developer Tools at AWS 2018.pdf
Developer Tools at AWS 2018.pdfDeveloper Tools at AWS 2018.pdf
Developer Tools at AWS 2018.pdf
 
SanDiego_DevOps_Meetup_9212016-v8
SanDiego_DevOps_Meetup_9212016-v8SanDiego_DevOps_Meetup_9212016-v8
SanDiego_DevOps_Meetup_9212016-v8
 
ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...
ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...
ENT201 A Tale of Two Pizzas: Accelerating Software Delivery with AWS Develope...
 
Announcing AWS CodeBuild - January 2017 Online Teck Talks
Announcing AWS CodeBuild - January 2017 Online Teck TalksAnnouncing AWS CodeBuild - January 2017 Online Teck Talks
Announcing AWS CodeBuild - January 2017 Online Teck Talks
 
Modernizing Testing as Apps Re-Architect
Modernizing Testing as Apps Re-ArchitectModernizing Testing as Apps Re-Architect
Modernizing Testing as Apps Re-Architect
 
Dev/Test scenarios in DevOps world
Dev/Test scenarios in DevOps worldDev/Test scenarios in DevOps world
Dev/Test scenarios in DevOps world
 
[WSO2Con EU 2017] Continuous Integration, Delivery and Deployment: Accelerate...
[WSO2Con EU 2017] Continuous Integration, Delivery and Deployment: Accelerate...[WSO2Con EU 2017] Continuous Integration, Delivery and Deployment: Accelerate...
[WSO2Con EU 2017] Continuous Integration, Delivery and Deployment: Accelerate...
 
What is DevOps?
What is DevOps?What is DevOps?
What is DevOps?
 
DBD 2414 - Iterative Web-Based Designer for Software Defined Environments (In...
DBD 2414 - Iterative Web-Based Designer for Software Defined Environments (In...DBD 2414 - Iterative Web-Based Designer for Software Defined Environments (In...
DBD 2414 - Iterative Web-Based Designer for Software Defined Environments (In...
 
Perforce on Tour 2015 - Optimising the Developer Pipeline: Deliver Faster & ...
Perforce on Tour 2015 -  Optimising the Developer Pipeline: Deliver Faster & ...Perforce on Tour 2015 -  Optimising the Developer Pipeline: Deliver Faster & ...
Perforce on Tour 2015 - Optimising the Developer Pipeline: Deliver Faster & ...
 

Recently uploaded

notes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptnotes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.ppt
MsecMca
 
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak HamilCara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Kandungan 087776558899
 
scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...
scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...
scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...
HenryBriggs2
 
Integrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - NeometrixIntegrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - Neometrix
Neometrix_Engineering_Pvt_Ltd
 
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
ssuser89054b
 
"Lesotho Leaps Forward: A Chronicle of Transformative Developments"
"Lesotho Leaps Forward: A Chronicle of Transformative Developments""Lesotho Leaps Forward: A Chronicle of Transformative Developments"
"Lesotho Leaps Forward: A Chronicle of Transformative Developments"
mphochane1998
 

Recently uploaded (20)

Minimum and Maximum Modes of microprocessor 8086
Minimum and Maximum Modes of microprocessor 8086Minimum and Maximum Modes of microprocessor 8086
Minimum and Maximum Modes of microprocessor 8086
 
Generative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPTGenerative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPT
 
Employee leave management system project.
Employee leave management system project.Employee leave management system project.
Employee leave management system project.
 
Engineering Drawing focus on projection of planes
Engineering Drawing focus on projection of planesEngineering Drawing focus on projection of planes
Engineering Drawing focus on projection of planes
 
Design For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the startDesign For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the start
 
notes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptnotes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.ppt
 
Tamil Call Girls Bhayandar WhatsApp +91-9930687706, Best Service
Tamil Call Girls Bhayandar WhatsApp +91-9930687706, Best ServiceTamil Call Girls Bhayandar WhatsApp +91-9930687706, Best Service
Tamil Call Girls Bhayandar WhatsApp +91-9930687706, Best Service
 
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak HamilCara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
Cara Menggugurkan Sperma Yang Masuk Rahim Biyar Tidak Hamil
 
A Study of Urban Area Plan for Pabna Municipality
A Study of Urban Area Plan for Pabna MunicipalityA Study of Urban Area Plan for Pabna Municipality
A Study of Urban Area Plan for Pabna Municipality
 
kiln thermal load.pptx kiln tgermal load
kiln thermal load.pptx kiln tgermal loadkiln thermal load.pptx kiln tgermal load
kiln thermal load.pptx kiln tgermal load
 
scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...
scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...
scipt v1.pptxcxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx...
 
Integrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - NeometrixIntegrated Test Rig For HTFE-25 - Neometrix
Integrated Test Rig For HTFE-25 - Neometrix
 
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
 
HOA1&2 - Module 3 - PREHISTORCI ARCHITECTURE OF KERALA.pptx
HOA1&2 - Module 3 - PREHISTORCI ARCHITECTURE OF KERALA.pptxHOA1&2 - Module 3 - PREHISTORCI ARCHITECTURE OF KERALA.pptx
HOA1&2 - Module 3 - PREHISTORCI ARCHITECTURE OF KERALA.pptx
 
Block diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.pptBlock diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.ppt
 
Work-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxWork-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptx
 
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
 
"Lesotho Leaps Forward: A Chronicle of Transformative Developments"
"Lesotho Leaps Forward: A Chronicle of Transformative Developments""Lesotho Leaps Forward: A Chronicle of Transformative Developments"
"Lesotho Leaps Forward: A Chronicle of Transformative Developments"
 
Double Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torqueDouble Revolving field theory-how the rotor develops torque
Double Revolving field theory-how the rotor develops torque
 
S1S2 B.Arch MGU - HOA1&2 Module 3 -Temple Architecture of Kerala.pptx
S1S2 B.Arch MGU - HOA1&2 Module 3 -Temple Architecture of Kerala.pptxS1S2 B.Arch MGU - HOA1&2 Module 3 -Temple Architecture of Kerala.pptx
S1S2 B.Arch MGU - HOA1&2 Module 3 -Temple Architecture of Kerala.pptx
 

DevOps for AI Apps

  • 1. DevOps for AI Apps Richin Jain, Software Engineer (@richinjain) Vivek Gupta, Data Scientist (@gkeviv)
  • 2. Agenda • Background • DevOps Introduction • Enterprise AI use case • Workflows • Traditional v/s AI App • Proposed approach • Pipeline
  • 3. Background • Over the course of engaging with various enterprise customers on their AI use case, this has been a common ask. • At one point in time Data Science was one off task where answer was given. • Now it has been integrated with real time applications along with retraining and A/B testing. • Because of this we have to relook at Data Science process and how it could be integrated with existing software stack. • Goal was to start looking at best practices from S/W engineering and how could it best applied here.
  • 4. DevOps brings together people, processes, and technology, automating software delivery to provide continuous value to your users. What is DevOps? Continuous Integration (CI) • Focuses on blending the work of individual developers together into a repository. • Each time you commit code, it’s automatically built and tested and bugs are detected faster. Continuous Deployment (CD) • Automate the entire process from code commit to production if your CI/CD tests are successful. Continuous Learning & Monitoring • Using CI/CD practices, paired with monitoring tools, safely deliver features to your customers as soon as they’re ready.
  • 6. Configure Code Build Test Package Deploy Monitor • Infrastructure as Code • Continuous Integration • Automated Testing • Continuous Deployment • Release Management • Load Testing & Auto-Scale • App Performance Monitoring DevOps Maturity
  • 7. Data Science • Experimentation • Modeling • Versioning • Lineage • Conversion • Export • Quantization • Inferencing • Retraining • A/B Testing A C T I V I T I E S • Need to solve ML problem quickly. • ML stack might be different from rest of the application stack. • Lots of glue code. • Testing accuracy of ML model. • ML code is not always version controlled. • Hard to reproduce models • Integrating model into application can take weeks • Need to re-write featurizing and scoring code multiple times (in different languages) • Want to start using customer data to build models • Hard to track breaking changes P A I N P O I N T S
  • 8. Enterprise AI use case • Contoso LLC has an image recognition scenario. Data Science team develops a state-of-the-art image recognition model. • Four ways it could be consumed • User upload the images to Contoso's website and get instant results. • User uploads several images or point to a folder and get results. • Native mobile app • Edge devices
  • 9. API Based model integration • Real time • Batch Embedded models • Native Apps • Edge devices
  • 10. Workflow: App DeveloperWorkflow: App DeveloperBasic Workflow: Software Engineer
  • 11. Workflow: Data ScientistWorkflow: App DeveloperBasic Workflow: Data Scientist
  • 12.
  • 18. Build Artifact Create Conda Environment Install Requirements Deploy to Test (create/update) Test environment (continuous deployment) Dev Artifact Create Conda Environment Install Requirements Deploy to Staging (update) Staging Environment (nightly, other services test here) Test Artifact Create Conda Environment Install Requirements Deploy to Prod (update) Prod environment (end of sprint)
  • 19. Get Source Create Conda Environment Install Requirements Convert model to other formats ONNX, CoreML, WinML (end of sprint, every time there is a new model) Get Source Create Conda Environment Install Requirements Retrain model on new data Retraining Pipeline (every night, or triggered on new data uploading to blob) It will run in a pre-prod environment, so it has access to production data, and wouldn’t be promoted unless it passes A/B tests against prod data. Data Validation A/B Testing Model Testing and Validation Model Management & Promotion
  • 20. • Core features of Azure ML service exposed through a Python SDK and CLI. • Easy and simple pip install • Makes CI/CD much simpler.
  • 21.
  • 22.
  • 23. • Best practices for architecting and managing an enterprise-ready AI application lifecycle. • Azure DevOps and Azure ML ease the adoption of DevOps by DS teams. • Adoption will increase the agility, quality and delivery of DS teams.

Editor's Notes

  1. https://blogs.msdn.microsoft.com/visualstudioalmrangers/2017/04/20/set-up-a-cicd-pipeline-to-run-automated-tests-efficiently/
  2. Microsoft DevOps site - https://www.microsoft.com/en-us/cloud-platform/development-operations Source - http://www.itproguy.com/devops-practices/
  3. Link to Google paper - https://ai.google/research/pubs/pub45742
  4. Link to Google paper - https://ai.google/research/pubs/pub45742
  5. https://azure.microsoft.com/en-us/blog/what-s-new-in-azure-machine-learning-service/