Slide 1
Slide 1 text
Will AI Assistant make developers
redundant?
Slide 2
Slide 2 text
AI is everywhere
Slide 3
Slide 3 text
AI at conferences
Slide 4
Slide 4 text
AI in products
Slide 5
Slide 5 text
No content
Slide 6
Slide 6 text
https://x.com/petergyang/status/1793480607198323196
Slide 7
Slide 7 text
https://spectrum.ieee.org/in-2016-microsofts-racist-chatbot-revealed-the-dangers-of-online-conversation
Slide 8
Slide 8 text
No content
Slide 9
Slide 9 text
No content
Slide 10
Slide 10 text
Bold claims
Slide 11
Slide 11 text
It will make us more productive!
Slide 12
Slide 12 text
It will take our jobs!
Slide 13
Slide 13 text
AGI is imminent!
Slide 14
Slide 14 text
No content
Slide 15
Slide 15 text
Will AI take our jobs?
Slide 16
Slide 16 text
Jobs have
disappeared before
https://hollycummins.com/will-ai-take-our-jobs/
Slide 17
Slide 17 text
The claim that tech
will take our jobs
is not new
Slide 18
Slide 18 text
• Low-code / no code
Slide 19
Slide 19 text
• Low-code / no code
• CASE tools / 4GL
Slide 20
Slide 20 text
• Low-code / no code
• CASE tools / 4GL
• COBOL
Slide 21
Slide 21 text
Increasing demand for devs
https://hollycummins.com/will-ai-take-our-jobs/
Slide 22
Slide 22 text
Jobs will change
Slide 23
Slide 23 text
Does it make us more productive?
Slide 24
Slide 24 text
Productivity
https://github.blog/2022-09-07-research-quantifying-github-copilots-impact-on-developer-productivity-and-happiness/
Slide 25
Slide 25 text
Productivity
https://github.blog/2023-06-13-survey-reveals-ais-impact-on-the-developer-experience/
Slide 26
Slide 26 text
Productivity
"Based on data from 250K+ developers in our global community, developers code 52 minutes per day —
about 4 hours and 21 minutes during a normal workweek from Monday to Friday."
"Code time is defined as time spent actively writing or editing code in an editor or IDE."
https://www.software.com/reports/code-time-report
Slide 27
Slide 27 text
Understanding code
"developers on average spend as much as 58% of their time
comprehending existing source code"
~ Felienne Hermans, The Programmer's Brain
https://www.felienne.com/book
Slide 28
Slide 28 text
Understanding and maintaining
“The majority of a developer’s time isn't writing
but understanding and maintaining existing code.“
https://codescene.com/hubfs/whitepapers/Refactoring-vs-Refuctoring-Advancing-the-state-of-AI-automated-code-improvements.pdf
Slide 29
Slide 29 text
Productivity boost
https://blog.jetbrains.com/ai/2024/04/developers-save-up-to-8-hours-per-week-with-jetbrains-ai-assistant/
Slide 30
Slide 30 text
Time Saved
https://blog.jetbrains.com/ai/2024/04/developers-save-up-to-8-hours-per-week-with-jetbrains-ai-assistant/
Slide 31
Slide 31 text
Productivity boost
https://blog.jetbrains.com/ai/2024/04/developers-save-up-to-8-hours-per-week-with-jetbrains-ai-assistant/
Slide 32
Slide 32 text
Top 5 Features
https://blog.jetbrains.com/ai/2024/04/developers-save-up-to-8-hours-per-week-with-jetbrains-ai-assistant/
Slide 33
Slide 33 text
Top 5 Features
Personal observation:
This does not include generating code!
https://blog.jetbrains.com/ai/2024/04/developers-save-up-to-8-hours-per-week-with-jetbrains-ai-assistant/
Slide 34
Slide 34 text
Developer Experience Boost
https://blog.jetbrains.com/ai/2024/04/developers-save-up-to-8-hours-per-week-with-jetbrains-ai-assistant/
Slide 35
Slide 35 text
Our approach to building AI Assistant into our products has been focused on two main aspects:
1. Weaving the AI assistance into the core IDE user workflows.
2. Building AI features that are deeply infused with an understanding of your code and its context,
which has always been a strong suit of JetBrains IDEs.
JetBrains AI Assistant
https://blog.jetbrains.com/blog/2023/12/06/introducing-jetbrains-ai-and-the-in-ide-ai-assistant/
Slide 36
Slide 36 text
Generate and improve commit
messages, explain commit(s),
resolve merge conflicts
Continue prompting,
add context & commands
Generate, explain, refactor code,
write tests & documentation
Ask questions,
edit & save prompts,
see context, use code
Explain runtime errors,
write commands in new terminal
Get help
with SQL
queries in
console
Slide 37
Slide 37 text
Chat
● LLMs are good at chat 😁
● We can
● Ask how to do things
● Ask follow up questions
Slide 38
Slide 38 text
Chat
● Integrated into the IDE
● Context aware
● Can call back to the IDE
Slide 39
Slide 39 text
Chat
● Ask about your project
Slide 40
Slide 40 text
Chat
● Ask about your project
● Ask VCS related questions
Slide 41
Slide 41 text
Explain commit(s)
● Summarize changes in commit(s)
Slide 42
Slide 42 text
Chat
● Can add additional context
Slide 43
Slide 43 text
Chat
● Can add additional context
● Tag files, symbols & more
Slide 44
Slide 44 text
Explain code
Selected code added to prompt
Slide 45
Slide 45 text
Explain RegExp fragment
Slide 46
Slide 46 text
Explain cron fragment
Slide 47
Slide 47 text
Explain Spring Data QL fragment
Slide 48
Slide 48 text
Code generation is important
—
Slide 49
Slide 49 text
Code completion
● Auto completion while typing
● with syntax highlighting
Slide 50
Slide 50 text
Code completion
Cloud completion Local completion
Slide 51
Slide 51 text
Code completion
macOS Windows / Linux
Accept suggestion ⇥ TAB
Accept line ⌘→ END
Accept word ⌥→ Ctrl+→
● Change the key used to accept suggestion
Slide 52
Slide 52 text
Code completion
● Shortcut for auto completion
Slide 53
Slide 53 text
Code completion
● Shortcut to generate code
● According to your prompt
Slide 54
Slide 54 text
Code completion
● Generate code in editor
● Can add a follow-up message
Slide 55
Slide 55 text
Code completion
● Note: Might need small fixes
● (Alt+Enter is your friend)
Slide 56
Slide 56 text
AI Actions
Generate Code...
Slide 57
Slide 57 text
AI Actions
Generate Code...
Slide 58
Slide 58 text
AI Actions
From Generate menu
Slide 59
Slide 59 text
Generate commit message
● In the commit tool window
● According to default prompt
Slide 60
Slide 60 text
Customize commit message generation
Slide 61
Slide 61 text
Customize commit message generation
Slide 62
Slide 62 text
Improve commit message
● Only for commits you haven't pushed!
Slide 63
Slide 63 text
Find problems
Slide 64
Slide 64 text
Find problems
In commit
Slide 65
Slide 65 text
Main point
Slide 66
Slide 66 text
Photography
EV = Exposure value
t = Exposure time
N = Focal number =
focal lenght / diameter of aperture
EV = log2
N2
t
Slide 67
Slide 67 text
Photography
● Let's assume we have some code
● The code is not perfect
Slide 68
Slide 68 text
Photography
● Explain code
Slide 69
Slide 69 text
Photography
● Integrated into the IDE
● Can use it from the editor; no tool-switching
● But doesn't always get the context
● Let's give it a hint
Slide 70
Slide 70 text
Photography
● Rename method ev to exposure
Slide 71
Slide 71 text
Photography
● Let's ask again
● We can use specific instructions...
Slide 72
Slide 72 text
Photography
● Side note: options to use the code
Slide 73
Slide 73 text
Photography
... or use predefined prompt:
• Suggest Refactoring
Slide 74
Slide 74 text
Photography
Side note: options to use the code,
includes Show Diff
Slide 75
Slide 75 text
Photography
Slide 76
Slide 76 text
Photography
● Note this refactor includes renaming
● Can also use Refactor > Rename
● (especially if variables are used outside this file)
Slide 77
Slide 77 text
Photography
● Ask for real world scenarios
● (Prompt engineering is a thing)
Slide 78
Slide 78 text
Photography
● Convert to JUnit 5 tests
● with expressive names
Slide 79
Slide 79 text
Photography
● including exceptions and edge cases
● (Unlike some developers...)
Slide 80
Slide 80 text
Photography
● Might need some small fixes
● Add JUnit 5
Slide 81
Slide 81 text
Photography
● Might need some small fixes
● Add JUnit 5
● Import statements
Slide 82
Slide 82 text
Photography
● The numbers don't add up
Slide 83
Slide 83 text
Photography
Large Language Models
... are not good at math
https://www.youtube.com/watch?v=0xENpeGTEZ0
Slide 84
Slide 84 text
Photography
● Get real world data
● Talk to some users
Slide 85
Slide 85 text
Photography
● AI Actions > Generate unit tests
Slide 86
Slide 86 text
Photography
● Tests generated in existing test file or new file
Slide 87
Slide 87 text
Photography
● Add additional specifications...
Slide 88
Slide 88 text
Photography
... for better results
Slide 89
Slide 89 text
Photography
These tests also fail..
(Remember: LLMs are not good at math)
Slide 90
Slide 90 text
Photography
But are generated in the right place
Slide 91
Slide 91 text
Photography
● You can specify how to generate unit tests
● Customize prompt in Settings:
● Tools > AI Assistant > Prompt Library
Slide 92
Slide 92 text
Resolve merge conflicts
—
Slide 93
Slide 93 text
Resolve merge conflicts
Slide 94
Slide 94 text
Resolve merge conflicts
Slide 95
Slide 95 text
New Terminal
● Currently in beta
● Enable in Settings > Tools > Terminal
Slide 96
Slide 96 text
Use AI Assistant in the New Terminal
● Ask AI Assistant to generate command
Slide 97
Slide 97 text
Use AI Assistant in the New Terminal
● Ask AI Assistant to generate command
Slide 98
Slide 98 text
Use AI Assistant in the New Terminal
● Ask AI Assistant to generate command
● And run the command
Slide 99
Slide 99 text
Use AI Assistant to generate queries
Slide 100
Slide 100 text
Use AI Assistant to generate queries
Slide 101
Slide 101 text
Use AI Assistant to generate queries
Slide 102
Slide 102 text
Use AI Assistant to generate queries
Slide 103
Slide 103 text
Use AI Assistant to generate queries
● Attach database schema
● Might still need small fixes
● IntelliJ IDEA has DB support
Slide 104
Slide 104 text
Use AI Assistant to generate queries
Slide 105
Slide 105 text
AI Assistant
● Can save time
● Boilerplate code / typing
● Tasks we don't like to do
● Understanding existing code!
Slide 106
Slide 106 text
Real world use case
https://blog.jetbrains.com/idea/2023/12/java-inspection-with-ai-assistant/
Slide 107
Slide 107 text
Real world use case
"I definitely saved some time here, probably 10–20 minutes."
"The whole task took about 35 minutes, including creating a YouTrack issue and taking the screenshots
for this blogpost. Not bad!"
https://blog.jetbrains.com/idea/2023/12/java-inspection-with-ai-assistant/
Slide 108
Slide 108 text
Some caveats
"The main problem I see for these kinds of tasks is that AI cannot learn (yet)."
"When AI is able to accumulate and use project-specific knowledge, it will become way more useful."
https://blog.jetbrains.com/idea/2023/12/java-inspection-with-ai-assistant/
Slide 109
Slide 109 text
Challenges
https://www.jetbrains.com/guide/ai/links/critical-thinking-in-an-ai-powered-world/
Slide 110
Slide 110 text
Challenges: Maths
● Computers are good at maths
● LLMs are good at language
Slide 111
Slide 111 text
Challenge: Timeliness
● Cutoff date
● Suggestions may be (out)dated
● Be mindful when using latest tech!
Slide 112
Slide 112 text
Secure code
https://www.jetbrains.com/guide/ai/links/critical-thinking-in-an-ai-powered-world/
Slide 113
Slide 113 text
Secure code
● Be critical in evaluating the AI assistants suggestions
● Use additional tools and techniques to verify the code
Slide 114
Slide 114 text
Code quality
● Code churn
● Code smells
https://arxiv.org/abs/2304.10778 & https://www.gitclear.com/coding_on_copilot_data_shows_ais_downward_pressure_on_code_quality
Slide 115
Slide 115 text
Automation bias
"Automation bias is the propensity for humans to favor
suggestions from automated decision-making
systems and to ignore contradictory information
made without automation, even if it is correct."
https://en.wikipedia.org/wiki/Automation_bias
Slide 116
Slide 116 text
Hallucinations
https://www.ibm.com/topics/ai-hallucinations
Slide 117
Slide 117 text
Preventing hallucinations
"removes hallucinations from the candidates suggested by LLMs, then further enhances and ranks
suggestions based on static analysis techniques from program slicing, and finally leverages the IDE to
execute refactorings correctly."
https://arxiv.org/pdf/2401.15298
Slide 118
Slide 118 text
Privacy
"But we can't send data to the cloud"
Slide 119
Slide 119 text
https://www.jetbrains.com/ai/enterprise/
JetBrains AI Enterprise
—
Slide 120
Slide 120 text
https://www.jetbrains.com/ai/enterprise/
JetBrains AI Enterprise
—
Slide 121
Slide 121 text
Full line code completion
Runs locally!
https://www.youtube.com/watch?v=DLBiJ5kYUFg
Slide 122
Slide 122 text
Does it make us more productive?
Slide 123
Slide 123 text
Jobs will change
Slide 124
Slide 124 text
Changes
https://x.com/karpathy/status/1617979122625712128
Slide 125
Slide 125 text
Changes (EAP)
https://x.com/hszanowski/status/1834175128000688280
Slide 126
Slide 126 text
To reiterate
Make sure you understand the code (an) AI Assistant gives you!
Slide 127
Slide 127 text
Does AI replace developers?
• Produce more code faster
Slide 128
Slide 128 text
Does AI replace developers?
• Produce more code faster -> more code to maintain
Slide 129
Slide 129 text
https://x.com/karpathy/status/1617979122625712128
Slide 130
Slide 130 text
Does AI replace developers?
• Produce more code faster -> more code to maintain
• Scanning tools / good dev practices more important than ever!
Slide 131
Slide 131 text
Does AI replace developers?
• Produce more code faster -> more code to maintain
• Scanning tools / good dev practices more important than ever!
• Our skillset might change; we need to adapt
Slide 132
Slide 132 text
Does AI replace developers?
• TL;DR: No
Slide 133
Slide 133 text
Does AI replace developers?
• TL;DR: No
• or at least, not yet (?)
Slide 134
Slide 134 text
Slides & more
https://maritvandijk.com/presentations/ai-assistant-developers/
Slide 135
Slide 135 text
https://www.jetbrains.com/guide/ai/