[Technologistra] Gemini Code Assist for Developer Productivity

The session demonstrated how Gemini Code Assist accelerates the development lifecycle and project timelines. It highlighted how AI tools can augment and automate workflows, including generating code, automating unit tests, and creating documentation. The demo featured a fun use case: writing a prompt for Gemini to create a website that randomly assigns participants to Hogwarts houses from Harry Potter, with the assigned house broadcasted via the LINE Messaging API.

Transcript

1. Gemini Code Assist for Developer Productivity Code Plan Build Deploy O perate M onitor Release Test
2. Punsiri Boonyakiat - Senior Data Engineer @ CJMORE - Speaker @ Women Tech Maker 2024 - Speaker @ LINE DEV CONFERENCE 2021 - Speaker @ Skoodio & DATA TH about Gemini Code Assistance in Data Engineering Linkedin: https://www.linkedin.com/in/punsiri-boonyakiat-13909786/ Introduction
3. Are you a Software Tester, Developer, Data Science , or Data Engineer ?
4. be 40% more productive? Would you like to
5. Agenda 01 What is Generative AI 02 How AI code-assisted boost productivity 03 Demo : AI-assisted Dev workflow 04 Summary
6. What is GenAI
8. Discriminative vs. Generative AI
9. Type of Generative AI based on Input Data
10. Type of Generative AI based on Input Data
11. Generative AI & Use-Case Overview
12. Interact with LLM - Prompting Technique
13. Challenges in Large Language Model Hallucination
14. How AI code-assist boost developer productivity
15. Example of Today’s Tech Landscape
16. AI? Source: @forrestbrazeal
18. The traditional developer workflow Define project goals, scope and requirements. Create software architecture, user interface, and system design. Write code, implement features, and integrate components. Conduct unit tests, integration tests, and end-to-end tests to ensure code quality and functionality. Release the application to staging or production environments. Continuously monitor the application's performance, user feedback, and identify areas for improvement. Plan Design Code Release Monitor Test
19. Code Assistants Landscape
20. The AI-assisted workflow
21. AI
22. 44% of developers are already using AI tools in their workflow Stack Overflow Source: Stack Overflow 2023 Developer Survey
23. Why you should consider using an AI coding assistant ?
24. Code answers Guided migration Autocomplete Code chat Find bugs Automate code generation Automating test generation Automate bug fixing Automate code migrations The AI-assisted developer workflow Augment Intelligent agents Automate Spectrum of possibilities
25. Gemini Code Assist
26. In this approach we create an anthemic opener showing real people using A.I. to do amazing things. This can be everyday people In this approach we create an anthemic opener showing real people using A.I. to do amazing things. This can be everyday people 97% Statistic Caption In this approach we create an anthemic opener showing real people using A.I. to do amazing things. This can be everyday people Assistance across multiple surfaces Gemini Advanced Gemini Code Assist (VSCode, JetBrains) Gemini in IDX Gemini in Android Studio Gemini in Colab Gemini in Google Cloud Chat In-editor Code Assist AI developer assistance
27. Right-click and select ‘Replace image’ gemini.google.com Gemini Advance (web based)
28. Gemini Code Assist in Cloud Shell Gemini Code Assist in Visual Studio Code
29. Available for multiple IDEs and developer surfaces, multiple language
30. The AI-assisted developer workflow Requirements analysis User story generation UX design assistance Architecture diagrams Design > Code Code generation Code autocompletion Code answers Code migrations Assistive debugging Test case generation Test data-sets Test automation Deploy insights Autofix issues Automation script writing Generate CI/CD pipelines Anomaly detection Feedback analysis AI-support tools Design Code Plan Test Release Monitor
31. DEMO: AI-Assisted Dev workflow
32. To Start: Enable Gemini for Google Cloud
33. To Start: Add Gemini Code Assist Extension in VSCode
34. Code answers Guided migration Autocomplete Code chat Find bugs Automate code generation Automating test generation Automate bug fixing Automate ocument generation The AI-assisted developer workflow Augment Intelligent agents Automate Spectrum of possibilities
35. Gemini คัดสรร https://github.com/punsiriboo/gemini-code-assist-for-dev
36. https://github.com/punsiriboo/gemini-code-assist-for-dev Prompt: สรางเว็บไซตแบบ Static ที่มีชองกรอกชื่อ จากนั้นสุมบานในแฮรรี่ พอตเตอรให และแสดงชื่อและบานของคุณ โดยใช HTML และ JavaScript ผลลัพธเปน ชือ และบอกวาอยูบานไหนโดยใช รูปจาก folder images
37. https://github.com/punsiriboo/gemini-code-assist-for-dev Prompt: # generate function to random house in Harry potter and boardcast using BroadcastRequest # send ImageMessage contain house png image from https://storage.googleapis.com/gemini-code-assist-beat/ # send TextMessage contain house name and house description
38. Summary
39. AI across the software development workflow Code Generation Code Summarization Unit test generation Code refactoring Code migration Vulnerability scanning Automated release planning Streamline git process Assisted documentation Inner loop Outer loop Code Plan Build Deploy O perate M onitor Release Test
40. AI-assisted: best practices Include enough context Evaluate the output Augment, don’t replace Include the full error message and any relevant code Include code or data example context Critically evaluate any suggested fixes or explanations AI augments your skills, but doesn’t replace them
41. Include enough context Evaluate the output Augment, don’t replace Include the full error message and any relevant code Critically evaluate any suggested fixes or explanations AI augments your skills, but doesn’t replace them AI-assisted: best practices
42. Include enough context Evaluate the output Augment, don’t replace Include the full error message and any relevant code Critically evaluate any suggested fixes or explanations AI augments your skills, but doesn’t replace them AI-assisted: best practices
43. Include enough context Evaluate the output Augment, don’t replace Include the full error message and any relevant code Include code or data example context Critically evaluate any suggested fixes or explanations AI augments your skills, but doesn’t replace them AI-assisted: best practices
44. Flow in Gemini-code Assist
45. Gemini Coding Assistance Utility
46. How Coding Assistance may shape your Dev Workflow ?
47. Balancing Between Generative AI Code Assist Benefit vs.Challenge
48. ฝากไปอ่าน ติดตาม บทความ medium กันได้นะค่ะ https://medium.com/@punsiriboonyakiat