SlideShare a Scribd company logo
Director of Engineering at GitHub
The future of developer tools
Too many repetitive operations
We use the developer community to help us
What is the problem?
AI tools for code
generation
We could participate with the right structure with people who care
deeply about developing AI in a way that is safe and is beneficial to
humanity.
The best defense is to empower as many people as possible to
have AI. If everyone has AI powers, then there's not any one person
or a small set of individuals who can have AI superpower.
GPT - Generative Pre-Trained
Transformer
• GPT - an innovation in the
Natural Language Processing
(NLP) space
• Takes an input such as a sentence
and tries to generate an
appropriate response.
• Unsupervised and Pre-trained
• A machine learning model that can look at
part of a sentence and predict the next
word.
• The GPT-2 was trained on a massive 40GB
dataset called WebText
• GPT2, is opened sourced
Local Machine
Tabnine
Sequential Text Prediction Model
• Has been known to be the
most advanced of its kind
• Can understand the
meaning of a sentence
and try to output a
meaningful sentence
• Public can use OpenAI
APIs to make use of the
GPT-3 model.
• Codexisa descendentofGPT-3designedtoperformonespecializedtask (transformingfunctiondescriptionsandsignaturesintosourcecode)withhigh
accuracy.
• Thedeeplearningmodeldoesnotunderstandprogramming.Likeall otherdeeplearning–basedlanguagemodels,Codexis capturingstatistical
correlationsbetweencodefragments.
Copilot
Reviews
• Big time-saver. It built out entire
React components for me.
• Copilot can autofill repetitive
code if it senses a pattern
• Besides, providing suggestions
regarding code completion, it is
also a very good spell detector.
Critics
• Copilots, KITe and TabNine Analyzes the code on the file context only
• Copilot uses smaller memory size only 12 billions parameters vs 175 billion on GPT-3
• Performance is reduced when the size of the model increases
• The models are relatively new they need to be trained
All the products are built to learn from our preferences and make better code
suggsetions. So the more we use them the better they will become.
Security risk - If adversary uploads malicious code in
GitHub in enough abundance and targeted for a
specific type of prompt, Codex or GPT-2 might pick up
those patterns during training and then output them in
response to user instructions.
Licensing -what happens when the tool reproduces
code snippets thar are licensed and under copyright
protection? GitHub has said there is 0.1 percent
chance of Copilot replicating the learned snippet of
code verbatim.
Vulnerabilities & Bugs - Code often contains bugs—and
so, given the vast quantity of unvetted code that
Copilot has processed, it is certain that the language
model will have learned from exploitable, buggy code.
DevOps and AI
operate together
• Code reviews
• Software testing
• Monitor systems
• Resource management
• Anomaly detection & AIOps
SPOTLIGHT IGNITE (10 MINUTES): THE FUTURE OF DEVELOPER TOOLS: FROM STACKOVERFLOW TO CO-PILOT - Meirav Feiler, Github

More Related Content

What's hot

GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021
GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021
GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021
William Caban
 
Marko Berković
Marko BerkovićMarko Berković
Marko Berković
CodeFest
 
The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...
The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...
The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...
Schlomo Schapiro
 
How open source is driving DevOps innovation: CloudOpen NA 2015
How open source is driving DevOps innovation: CloudOpen NA 2015How open source is driving DevOps innovation: CloudOpen NA 2015
How open source is driving DevOps innovation: CloudOpen NA 2015
Gordon Haff
 
Data Science Challenges in Personal Program Analysis
Data Science Challenges in Personal Program AnalysisData Science Challenges in Personal Program Analysis
Data Science Challenges in Personal Program Analysis
Work-Bench
 
Pentaho 8 Reporting for Java Developers - Because details matter
Pentaho 8 Reporting for Java Developers - Because details matterPentaho 8 Reporting for Java Developers - Because details matter
Pentaho 8 Reporting for Java Developers - Because details matter
Francesco Corti
 
Scala from the Trenches
Scala from the Trenches Scala from the Trenches
Scala from the Trenches
Kfir Bloch
 
Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...
Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...
Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...
inovex GmbH
 
It's all about feedback - code review as a great tool in the agile toolbox
It's all about feedback - code review as a great tool in the agile toolboxIt's all about feedback - code review as a great tool in the agile toolbox
It's all about feedback - code review as a great tool in the agile toolbox
Stefan Lay
 
Scala from the Trenches - Java One 2016
Scala from the Trenches - Java One 2016Scala from the Trenches - Java One 2016
Scala from the Trenches - Java One 2016
Kfir Bloch
 
Git in the Enterprise: How to succeed at DevOps using Git and a monorepo
Git in the Enterprise: How to succeed at DevOps using Git and a monorepoGit in the Enterprise: How to succeed at DevOps using Git and a monorepo
Git in the Enterprise: How to succeed at DevOps using Git and a monorepo
Gina Bustos
 
Scaling Analysis Responsibly
Scaling Analysis ResponsiblyScaling Analysis Responsibly
Scaling Analysis Responsibly
Work-Bench
 
ATAGTR2017 Expanding test horizons with Robot Framework
ATAGTR2017 Expanding test horizons with Robot FrameworkATAGTR2017 Expanding test horizons with Robot Framework
ATAGTR2017 Expanding test horizons with Robot Framework
Agile Testing Alliance
 
ESE 2010: Using Git in Eclipse
ESE 2010: Using Git in EclipseESE 2010: Using Git in Eclipse
ESE 2010: Using Git in Eclipse
Chris Aniszczyk
 
What's New in GitLab and Software Development Trends
What's New in GitLab and Software Development TrendsWhat's New in GitLab and Software Development Trends
What's New in GitLab and Software Development Trends
Noa Harel
 
DevSecCon Boston2018 - advanced mobile security automation with bdd
DevSecCon Boston2018 - advanced mobile security automation with bddDevSecCon Boston2018 - advanced mobile security automation with bdd
DevSecCon Boston2018 - advanced mobile security automation with bdd
Davide Cioccia
 
Case Study: Migration to GitLab (from Bitbucket) at AppsFlyer
Case Study: Migration to GitLab (from Bitbucket) at AppsFlyerCase Study: Migration to GitLab (from Bitbucket) at AppsFlyer
Case Study: Migration to GitLab (from Bitbucket) at AppsFlyer
Noa Harel
 
Code Refactoring or Rewrite: How to Properly Dispose of Legacy Code
Code Refactoring or Rewrite: How to Properly Dispose of Legacy CodeCode Refactoring or Rewrite: How to Properly Dispose of Legacy Code
Code Refactoring or Rewrite: How to Properly Dispose of Legacy Code
Roman Labunsky
 
Attacking and defending GraphQL applications: a hands-on approach
 Attacking and defending GraphQL applications: a hands-on approach Attacking and defending GraphQL applications: a hands-on approach
Attacking and defending GraphQL applications: a hands-on approach
Davide Cioccia
 
Building A Distributed Build System at Google Scale (StrangeLoop 2016)
Building A Distributed Build System at Google Scale (StrangeLoop 2016)Building A Distributed Build System at Google Scale (StrangeLoop 2016)
Building A Distributed Build System at Google Scale (StrangeLoop 2016)
Aysylu Greenberg
 

What's hot (20)

GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021
GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021
GitOps, Driving NGN Operations Teams 211127 #kcdgt 2021
 
Marko Berković
Marko BerkovićMarko Berković
Marko Berković
 
The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...
The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...
The Role of GitOps in IT-Strategy - November 2021 - Schlomo Schapiro - Contin...
 
How open source is driving DevOps innovation: CloudOpen NA 2015
How open source is driving DevOps innovation: CloudOpen NA 2015How open source is driving DevOps innovation: CloudOpen NA 2015
How open source is driving DevOps innovation: CloudOpen NA 2015
 
Data Science Challenges in Personal Program Analysis
Data Science Challenges in Personal Program AnalysisData Science Challenges in Personal Program Analysis
Data Science Challenges in Personal Program Analysis
 
Pentaho 8 Reporting for Java Developers - Because details matter
Pentaho 8 Reporting for Java Developers - Because details matterPentaho 8 Reporting for Java Developers - Because details matter
Pentaho 8 Reporting for Java Developers - Because details matter
 
Scala from the Trenches
Scala from the Trenches Scala from the Trenches
Scala from the Trenches
 
Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...
Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...
Sprachsteuerung mit dem Google Assistant – Add a new User Interface to your P...
 
It's all about feedback - code review as a great tool in the agile toolbox
It's all about feedback - code review as a great tool in the agile toolboxIt's all about feedback - code review as a great tool in the agile toolbox
It's all about feedback - code review as a great tool in the agile toolbox
 
Scala from the Trenches - Java One 2016
Scala from the Trenches - Java One 2016Scala from the Trenches - Java One 2016
Scala from the Trenches - Java One 2016
 
Git in the Enterprise: How to succeed at DevOps using Git and a monorepo
Git in the Enterprise: How to succeed at DevOps using Git and a monorepoGit in the Enterprise: How to succeed at DevOps using Git and a monorepo
Git in the Enterprise: How to succeed at DevOps using Git and a monorepo
 
Scaling Analysis Responsibly
Scaling Analysis ResponsiblyScaling Analysis Responsibly
Scaling Analysis Responsibly
 
ATAGTR2017 Expanding test horizons with Robot Framework
ATAGTR2017 Expanding test horizons with Robot FrameworkATAGTR2017 Expanding test horizons with Robot Framework
ATAGTR2017 Expanding test horizons with Robot Framework
 
ESE 2010: Using Git in Eclipse
ESE 2010: Using Git in EclipseESE 2010: Using Git in Eclipse
ESE 2010: Using Git in Eclipse
 
What's New in GitLab and Software Development Trends
What's New in GitLab and Software Development TrendsWhat's New in GitLab and Software Development Trends
What's New in GitLab and Software Development Trends
 
DevSecCon Boston2018 - advanced mobile security automation with bdd
DevSecCon Boston2018 - advanced mobile security automation with bddDevSecCon Boston2018 - advanced mobile security automation with bdd
DevSecCon Boston2018 - advanced mobile security automation with bdd
 
Case Study: Migration to GitLab (from Bitbucket) at AppsFlyer
Case Study: Migration to GitLab (from Bitbucket) at AppsFlyerCase Study: Migration to GitLab (from Bitbucket) at AppsFlyer
Case Study: Migration to GitLab (from Bitbucket) at AppsFlyer
 
Code Refactoring or Rewrite: How to Properly Dispose of Legacy Code
Code Refactoring or Rewrite: How to Properly Dispose of Legacy CodeCode Refactoring or Rewrite: How to Properly Dispose of Legacy Code
Code Refactoring or Rewrite: How to Properly Dispose of Legacy Code
 
Attacking and defending GraphQL applications: a hands-on approach
 Attacking and defending GraphQL applications: a hands-on approach Attacking and defending GraphQL applications: a hands-on approach
Attacking and defending GraphQL applications: a hands-on approach
 
Building A Distributed Build System at Google Scale (StrangeLoop 2016)
Building A Distributed Build System at Google Scale (StrangeLoop 2016)Building A Distributed Build System at Google Scale (StrangeLoop 2016)
Building A Distributed Build System at Google Scale (StrangeLoop 2016)
 

More from DevOpsDays Tel Aviv

YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...
YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...
YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...
DevOpsDays Tel Aviv
 
GRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, Salto
GRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, SaltoGRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, Salto
GRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, Salto
DevOpsDays Tel Aviv
 
MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...
MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...
MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...
DevOpsDays Tel Aviv
 
THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...
THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...
THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...
DevOpsDays Tel Aviv
 
PRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDog
PRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDogPRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDog
PRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDog
DevOpsDays Tel Aviv
 
NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...
NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...
NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...
DevOpsDays Tel Aviv
 
(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG
(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG
(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG
DevOpsDays Tel Aviv
 
BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...
BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...
BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...
DevOpsDays Tel Aviv
 
THE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider Security
THE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider SecurityTHE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider Security
THE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider Security
DevOpsDays Tel Aviv
 
THE PLEASURES OF ON-PREM, TOMER GABEL
THE PLEASURES OF ON-PREM, TOMER GABELTHE PLEASURES OF ON-PREM, TOMER GABEL
THE PLEASURES OF ON-PREM, TOMER GABEL
DevOpsDays Tel Aviv
 
CONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPack
CONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPackCONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPack
CONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPack
DevOpsDays Tel Aviv
 
SOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, Develeap
SOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, DeveleapSOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, Develeap
SOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, Develeap
DevOpsDays Tel Aviv
 
OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...
OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...
OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...
DevOpsDays Tel Aviv
 
HOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKH
HOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKHHOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKH
HOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKH
DevOpsDays Tel Aviv
 
FLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, Icinga
FLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, IcingaFLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, Icinga
FLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, Icinga
DevOpsDays Tel Aviv
 
(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY
(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY
(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY
DevOpsDays Tel Aviv
 
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.ioSLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
DevOpsDays Tel Aviv
 
ONBOARDING IN LOCKDOWN, HILA FOX, Augury
ONBOARDING IN LOCKDOWN, HILA FOX, AuguryONBOARDING IN LOCKDOWN, HILA FOX, Augury
ONBOARDING IN LOCKDOWN, HILA FOX, Augury
DevOpsDays Tel Aviv
 
DON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, Firefly
DON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, FireflyDON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, Firefly
DON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, Firefly
DevOpsDays Tel Aviv
 
KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...
KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...
KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...
DevOpsDays Tel Aviv
 

More from DevOpsDays Tel Aviv (20)

YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...
YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...
YOUR OPEN SOURCE PROJECT IS LIKE A STARTUP, TREAT IT LIKE ONE, EYAR ZILBERMAN...
 
GRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, Salto
GRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, SaltoGRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, Salto
GRAPHQL TO THE RES(T)CUE, ELLA SHARAKANSKI, Salto
 
MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...
MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...
MICROSERVICES ABOVE THE CLOUD - DESIGNING THE INTERNATIONAL SPACE STATION FOR...
 
THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...
THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...
THE (IR)RATIONAL INCIDENT RESPONSE: HOW PSYCHOLOGICAL BIASES AFFECT INCIDENT ...
 
PRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDog
PRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDogPRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDog
PRINCIPLES OF OBSERVABILITY // DANIEL MAHER, DataDog
 
NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...
NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...
NUDGE AND SLUDGE: DRIVING SECURITY WITH DESIGN // J. WOLFGANG GOERLICH, Duo S...
 
(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG
(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG
(Ignite) TAKE A HIKE: PREVENTING BATTERY CORROSION - LEAH VOGEL, CHEGG
 
BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...
BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...
BUILDING A DR PLAN FOR YOUR CLOUD INFRASTRUCTURE FROM THE GROUND UP, MOSHE BE...
 
THE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider Security
THE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider SecurityTHE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider Security
THE THREE DISCIPLINES OF CI/CD SECURITY, DANIEL KRIVELEVICH, Cider Security
 
THE PLEASURES OF ON-PREM, TOMER GABEL
THE PLEASURES OF ON-PREM, TOMER GABELTHE PLEASURES OF ON-PREM, TOMER GABEL
THE PLEASURES OF ON-PREM, TOMER GABEL
 
CONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPack
CONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPackCONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPack
CONFIGURATION MANAGEMENT IN THE CLOUD NATIVE ERA, SHAHAR MINTZ, EggPack
 
SOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, Develeap
SOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, DeveleapSOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, Develeap
SOLVING THE DEVOPS CRISIS, ONE PERSON AT A TIME, CHRISTINA BABITSKI, Develeap
 
OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...
OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...
OPTIMIZING PERFORMANCE USING CONTINUOUS PRODUCTION PROFILING ,YONATAN GOLDSCH...
 
HOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKH
HOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKHHOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKH
HOW TO SCALE YOUR ONCALL OPERATION, AND SURVIVE TO TELL, ANTON DRUKH
 
FLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, Icinga
FLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, IcingaFLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, Icinga
FLYING BLIND - ACCESSIBILITY IN MONITORING, FEU MOUREK, Icinga
 
(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY
(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY
(Ignite) WHAT'S BURNING THROUGH YOUR CLOUD BILL - GIL BAHAT, CIDER SECURITY
 
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.ioSLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
SLO DRIVEN DEVELOPMENT, ALON NATIV, Tomorrow.io
 
ONBOARDING IN LOCKDOWN, HILA FOX, Augury
ONBOARDING IN LOCKDOWN, HILA FOX, AuguryONBOARDING IN LOCKDOWN, HILA FOX, Augury
ONBOARDING IN LOCKDOWN, HILA FOX, Augury
 
DON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, Firefly
DON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, FireflyDON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, Firefly
DON'T PANIC: GETTING YOUR INFRASTRUCTURE DRIFT UNDER CONTROL, ERAN BIBI, Firefly
 
KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...
KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...
KEYNOTE | WHAT'S COMING IN THE NEXT 10 YEARS OF DEVOPS? // ELLEN CHISA, bolds...
 

Recently uploaded

National Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practicesNational Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practices
Quotidiano Piemontese
 
Video Streaming: Then, Now, and in the Future
Video Streaming: Then, Now, and in the FutureVideo Streaming: Then, Now, and in the Future
Video Streaming: Then, Now, and in the Future
Alpen-Adria-Universität
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Albert Hoitingh
 
Microsoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdfMicrosoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdf
Uni Systems S.M.S.A.
 
Monitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR EventsMonitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR Events
Ana-Maria Mihalceanu
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
James Anderson
 
UiPath Test Automation using UiPath Test Suite series, part 5
UiPath Test Automation using UiPath Test Suite series, part 5UiPath Test Automation using UiPath Test Suite series, part 5
UiPath Test Automation using UiPath Test Suite series, part 5
DianaGray10
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform Engineering
Jemma Hussein Allen
 
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdfFIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance
 
Secstrike : Reverse Engineering & Pwnable tools for CTF.pptx
Secstrike : Reverse Engineering & Pwnable tools for CTF.pptxSecstrike : Reverse Engineering & Pwnable tools for CTF.pptx
Secstrike : Reverse Engineering & Pwnable tools for CTF.pptx
nkrafacyberclub
 
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdfFIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
Kari Kakkonen
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
Safe Software
 
20240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 202420240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 2024
Matthew Sinclair
 
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
Neo4j
 
Climate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing DaysClimate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing Days
Kari Kakkonen
 
The Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesThe Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and Sales
Laura Byrne
 
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
DanBrown980551
 
Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1
DianaGray10
 

Recently uploaded (20)

National Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practicesNational Security Agency - NSA mobile device best practices
National Security Agency - NSA mobile device best practices
 
Video Streaming: Then, Now, and in the Future
Video Streaming: Then, Now, and in the FutureVideo Streaming: Then, Now, and in the Future
Video Streaming: Then, Now, and in the Future
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
 
Microsoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdfMicrosoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdf
 
Monitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR EventsMonitoring Java Application Security with JDK Tools and JFR Events
Monitoring Java Application Security with JDK Tools and JFR Events
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
 
UiPath Test Automation using UiPath Test Suite series, part 5
UiPath Test Automation using UiPath Test Suite series, part 5UiPath Test Automation using UiPath Test Suite series, part 5
UiPath Test Automation using UiPath Test Suite series, part 5
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform Engineering
 
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdfFIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
 
Secstrike : Reverse Engineering & Pwnable tools for CTF.pptx
Secstrike : Reverse Engineering & Pwnable tools for CTF.pptxSecstrike : Reverse Engineering & Pwnable tools for CTF.pptx
Secstrike : Reverse Engineering & Pwnable tools for CTF.pptx
 
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdfFIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
FIDO Alliance Osaka Seminar: FIDO Security Aspects.pdf
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
 
20240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 202420240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 2024
 
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
 
Climate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing DaysClimate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing Days
 
The Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesThe Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and Sales
 
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
LF Energy Webinar: Electrical Grid Modelling and Simulation Through PowSyBl -...
 
Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1
 

SPOTLIGHT IGNITE (10 MINUTES): THE FUTURE OF DEVELOPER TOOLS: FROM STACKOVERFLOW TO CO-PILOT - Meirav Feiler, Github

  • 2. The future of developer tools
  • 3. Too many repetitive operations
  • 4. We use the developer community to help us
  • 5. What is the problem?
  • 6. AI tools for code generation
  • 7. We could participate with the right structure with people who care deeply about developing AI in a way that is safe and is beneficial to humanity. The best defense is to empower as many people as possible to have AI. If everyone has AI powers, then there's not any one person or a small set of individuals who can have AI superpower.
  • 8. GPT - Generative Pre-Trained Transformer • GPT - an innovation in the Natural Language Processing (NLP) space • Takes an input such as a sentence and tries to generate an appropriate response. • Unsupervised and Pre-trained
  • 9. • A machine learning model that can look at part of a sentence and predict the next word. • The GPT-2 was trained on a massive 40GB dataset called WebText • GPT2, is opened sourced
  • 11.
  • 12. Sequential Text Prediction Model • Has been known to be the most advanced of its kind • Can understand the meaning of a sentence and try to output a meaningful sentence • Public can use OpenAI APIs to make use of the GPT-3 model.
  • 13. • Codexisa descendentofGPT-3designedtoperformonespecializedtask (transformingfunctiondescriptionsandsignaturesintosourcecode)withhigh accuracy. • Thedeeplearningmodeldoesnotunderstandprogramming.Likeall otherdeeplearning–basedlanguagemodels,Codexis capturingstatistical correlationsbetweencodefragments.
  • 15.
  • 16. Reviews • Big time-saver. It built out entire React components for me. • Copilot can autofill repetitive code if it senses a pattern • Besides, providing suggestions regarding code completion, it is also a very good spell detector.
  • 18. • Copilots, KITe and TabNine Analyzes the code on the file context only • Copilot uses smaller memory size only 12 billions parameters vs 175 billion on GPT-3 • Performance is reduced when the size of the model increases • The models are relatively new they need to be trained
  • 19. All the products are built to learn from our preferences and make better code suggsetions. So the more we use them the better they will become.
  • 20. Security risk - If adversary uploads malicious code in GitHub in enough abundance and targeted for a specific type of prompt, Codex or GPT-2 might pick up those patterns during training and then output them in response to user instructions. Licensing -what happens when the tool reproduces code snippets thar are licensed and under copyright protection? GitHub has said there is 0.1 percent chance of Copilot replicating the learned snippet of code verbatim. Vulnerabilities & Bugs - Code often contains bugs—and so, given the vast quantity of unvetted code that Copilot has processed, it is certain that the language model will have learned from exploitable, buggy code.
  • 21. DevOps and AI operate together • Code reviews • Software testing • Monitor systems • Resource management • Anomaly detection & AIOps

Editor's Notes

  1. I am very excited to be here today.  Let me first introduce myself.My name is Meirav and I am a Director of engineering at GitHub owning npm the public registry of node packages, today I am not going to talk about the past or what my teams has been doing even though it's pretty interesting,
  2. Today I am going to talk about the future. Specifically the future of developer tools. But Before I talk about the future we should probably talk about the present and what is the problem with it.
  3. As developers we have to do a lot of repetitive and sometimes even boring tasks, like creating authentication models, http clients or implementing CRUD operations. As developers we don't want to invent the wheel every time over and over again so
  4. We go to the community to help us, there are a bunch of platforms for it, the most known ones are stack overflow. With very basic string syntax we search for what we are looking for and usually the search engine will give us back some Stack Overflow solution. In most cases we will copy and paste it (or some similar version of it) to our IDE.
  5. This process is time consuming, error prone and distracting, we lose our focus and context every time we leave the IDE to the browser and make decisions on our code that might be risky. However, relying on the knowledge of the developer community is important and very helpful for all of us. So Instead of searching on the web for solutions, it seems that integrating something similar as stack overflow inside our IDE will make developers more efficient and less likely to make mistakes.
  6. So with the continued growth of technology, prediction tools such as Intellij and AI systems a new line of developer tools emerges such as CoPilot , Kite and TabNine which I think are going to shape the future of developers. These tools have an AI engine that is able to give code suggestions for whole lines or entire functions right inside the IDE based on simple sentences. Today I am going to share with you a bit more details about these tools and how they work and how we can all benefit from them.
  7. I used a lot of big terms such as gpt-2 gpt -3 codex aמd more let me explain a bit more what do they mean..  OpenAI is an AI research lab company.  They are the ones who  Created the generative pre-training (GPT) language models.  They Deliver API’s that can provide a general-purpose , you input some text , and the model will generate a text completion that attempts to match whatever context or pattern you gave it. The reason they develop such APIs is because of their vision of making AI accessible for everyone, they believe that if everyone has the power to use AI it will ensure that not any one person or a small set of individuals can have AI superpower.
  8. GPT generative pre-training transformers is an innovation in the Natural Language Processing (NLP) NLP are models which aim to make computers understand the unstructured language human speaks and retrieve meaningful pieces of information from it. The groundbreaking change with GPT is that unlike NLP previous models  it wasn't trained for a specific task it’s general and it’s using the unsupervised approach for the machine learning algorithm.  There are two types of Machine Learning algorithms: Supervised and Unsupervised. Supervised learning includes all those algorithms that must need labeled data and can verify what they have learned. Or in other words is able to identify if the answer is right or wrong. Supervised learning isn’t something humans dont really do. Rather, most of the time, we collect knowledge based on our experience, or intuitions. That’s what roughly you can regard as unsupervised learning. The algorithm is not provided with any pre-assigned labels or scores for the training data. In unsupervised learning, an AI system will group unsorted information according to similarities and differences even though there are no categories provided.
  9. GPT-2  is A machine learning model that can look at part of a sentence and predict the next word. The most famous language models we all know are smartphone keyboards that suggest the next word based on what you’ve currently typed.  GPT-2 is open sourced and a direct scale-up of GPT, with more than 10X the parameters and trained on more than 10X the amount of data.The model uses 1.5 billion parameters and trained on a dataset of 8 million web pages.It is trained with a simple objective: predict the next word, given all of the previous words within some text.
  10. First I will talk about Tebnine, their first version of their product was published in 2018. It works with 21 IDEs and 30 programming languages. Their AI engine called Deep TabNine is based on a GPT-2 model ,which I will explain a bit later what it is, but it short it can predict the next word, given all of the previous words within some text. . Deep TabNine was trained on 2 millions of GitHub’s open source repositories.  As you can see in this diagram the Plugin listens to the keyboard and uses the file you are working on as the context for the input sending the information to the Deep Tabnine model, which suggests solutions. The plugin registers the suggestion you choose in order to improve its suggestion for the next time.  Tabnine runs locally on your machine by installing its models once you register.  The pros of having everything locally keeps your code secure and the suggestion mechanism becomes more suited to your preferences.  However its known that the gpt-2 model requires  a lot of computing power so if you dont have a strong machine you might feel that the plugin is slow or not responsive, another downside is that with local configuration you are losing the tool’s  improvements coming from public usage.  Tabnine prefer the local configuration because this way your code never leaves the local machine. However they recently published a cloud version of their tools but you need to opt in to use it.
  11. This is a short demo of Tabnine using Type secript. As you can see it looks very similar to any autocomplete plugin. However you can see that the suggestions are using past context such as parameter name. 
  12. GPT-3 Could Be Called a Sequential Text Prediction Model.  Its the 3rd version release and the upgraded version of GPT-2. Version 3 takes the GPT model to a whole new level as it’s trained on 175 billion parameters (which is over 10x the size of, GPT-2).  GPT-3 can now go further with tasks such as answering questions, writing essays, text summarization, language translation, and generating computer code.  The algorithmic structure of GPT-3 has been known to be the most advanced of its kind thanks to the vast amount of data used to pre-train it.  To generate sentences after taking an input, GPT-3 uses the field of semantics to understand the meaning of language and try to output a meaningful sentence for the user. The model does not learn what is correct or incorrect as it does not use labelled or supervised.
  13. OpenAI Codex is a direct descendant of GPT-3 that has been trained  for programming tasks. Its significantly more capable than GPT-3 in code generation, , because it was trained on a data set that includes a much larger concentration of public source code. Due to memory and data limitations codex uses only 12 billion parameters not like the original GPT-3 model who uses 175 billion parameters.Making it less accurate then GPT-3.  
  14. GitHub recently launched Co-Pilot, which is the newest AI auto completion tool.  It currently works only on 3 IDES and 2 programming languages and is in a beta phase. What's interesting with Co-Pilot that its based on GPT-3  which can generate sequences of text not only single word like GPT-2.  Similarly to Tabnine the Copilot plugin communicates with the IDE sending the context of the current file to the AI model Codex. The model responses with text suggestions that are then displayed in the monitor. Once a suggestion has been chosen the plugin will send back telemetry to improve the suggestions in the future.  Unlike Tabnine , Codex  is hosted on the cloud and shared with all users, making the community a significant player in their product.
  15. Here is a short demo of copilot and how its able by reading a comment to suggest a full function.
  16. There are a lot of positive  reviews on how these tools are efficient. Let me read them. 
  17. There are also some critics…. 
  18. There are many reasons why these products are still not that great , here are a few.  For now all products work on the context of a single file, that does not work well on big projects. Where for instance you define functions on different files.  It was also found that  GPT’s models efficiently decrease  when we increase the number parameters, so adding parameters  might give us more accurate results but it will take more time for getting the result. Also we need more data to train the models for more parameters which willrequire to scan private repositories, not an easy task . If you recall I mentioned the codex was trained on 12 Billion parameters not 175 Billion like the GPT-3 original model.  The tools are relatively new, they need to be trained and used to become better.  For most of these reasons there isn’t much that we can actually do , we just need to wait for the next version improvements , however
  19. We can help train products by using them.  This way the developers community can help shape the future of these tools.
  20. Now, I am sure that some of you might be asking yourself, so shoul I be looking for a new job? Are these tools going to replace us developers?  My short and simple answer is NO. It might change it though.  Replacing developers isn’t the aim or something that I think would ever happen. Currently the engine can’t understand a real-world problem, plan a solution, build it and show it off to the world — these tasks are what developers are good at and that will probably won’t change.  However, with the power of the developer community , tools like Copilot and Tabnine can be a game changer in the programming industry , not by stealing jobs, but by making developers more productive. We’ve been improving developers’ experience (code editors, debugging tools, etc.) since the last century, and now with the rise of AI technology, we can expect the creation of many more tools using it. New technologies usually create new jobs!
  21. These really cool products also come with challenges yet to be solved. The most straightforward one is the security concern. A sophisticated attacker can target  malicious code to a specific prompt  that can be picked up by the models, causing users to use malwared code.  Licensing issues, what happens when the tool reproduces code snippets that are licensed and under copyright protection?  Vulnerabilities & Bugs - Code often contains bugs and given the vast quantity of unvetted code that Copilot and Tabnine has processed, it is certain that the language model has learned from exploitable, buggy code and might suggest it. 
  22. So what can we do about it?  Don’t blindly accept the tool’s recommendations, same as you will never copy paste blindly a solution from stackoverflow.  Don’t let unexperienced developers use the tool without proper guidance , these tools are not ways to learn how to code properly.   Use Automated tools for vulnerabilities scans, for instance depndabot or component government  Improve your code reviewing skills as its required before accepting any suggestion from these tools.
  23. Today I spoke briefly on one part of the development cycle but I believe that the future of DevOps will be AI-driven. Humans are not equipped to handle the massive volumes of data and computing in daily operations of high traffic products, artificial intelligence will become a critical tool for computing analyzing and transforming how teams develop, deliver, deploy, and manage applications.  DevOps and AI can become interdependent.  DevOps is a business-driven approach to deliver software, and AI is the technology that can be integrated into the system for enhanced functionality. With the help of AI, DevOps teams can test, code, release, and monitor software more efficiently.
  24. Now that you know about these tools and the algorithms it uses,  I hope you will not be afraid of AI being an integrated part of the developer's life cycle.  rather excited about it and maybe you will even take action and help shape it.  Thank you!