Krakow, Poland, 19 - 21 June 2024
How much of your code can LLMs really write? Gemini Code Assist Case Study.
Tools-in-Action (INTERMEDIATE level)
Room 2
Large Language Models (LLMs) have demonstrated remarkable capabilities in various domains, including code generation. This presentation explores the potential of LLMs to revolutionize software development by examining the extent to which they can automate the coding process. Using Gemini Code Assist, a cutting-edge LLM-powered coding assistant, we delve into real-world case studies to assess its effectiveness in generating functional, efficient, and secure code.
We analyze the strengths and limitations of Gemini Code Assist in different scenarios, ranging from simple tasks to complex projects. Through a series of practical examples, we showcase how Gemini Code Assist can significantly enhance developer productivity by automating repetitive tasks, providing intelligent code suggestions, and even generating entire code snippets.
However, we also acknowledge the challenges and potential risks associated with relying solely on LLMs for code generation. We discuss the importance of human oversight, the need for robust validation mechanisms, and the ethical considerations surrounding the use of AI in software development.
By the end of this presentation, attendees will gain a deeper understanding of the current state of LLMs in code generation, the potential benefits and drawbacks of using tools like Gemini Code Assist, and the future trajectory of AI-powered software development.
We analyze the strengths and limitations of Gemini Code Assist in different scenarios, ranging from simple tasks to complex projects. Through a series of practical examples, we showcase how Gemini Code Assist can significantly enhance developer productivity by automating repetitive tasks, providing intelligent code suggestions, and even generating entire code snippets.
However, we also acknowledge the challenges and potential risks associated with relying solely on LLMs for code generation. We discuss the importance of human oversight, the need for robust validation mechanisms, and the ethical considerations surrounding the use of AI in software development.
By the end of this presentation, attendees will gain a deeper understanding of the current state of LLMs in code generation, the potential benefits and drawbacks of using tools like Gemini Code Assist, and the future trajectory of AI-powered software development.
Mateusz Zięba
Google
Mateusz is a seasoned software engineer with over a decade of experience at Google. He has contributed to major projects such as YouTube and Android TV, showcasing his expertise in building scalable and high-impact software systems. Currently, Mateusz is a staff software engineer on the Gemini Code Assist team, where he is at the forefront of developing AI-powered tools to revolutionize software development.
Beyond his professional achievements, Mateusz is an outdoor enthusiast who enjoys hiking, skiing, and running. He also has a passion for fish keeping, which reflects his meticulous nature and love for detail. Mateusz is fluent in Java , Go and Python, and he is constantly exploring new technologies and approaches to improve his craft.
Beyond his professional achievements, Mateusz is an outdoor enthusiast who enjoys hiking, skiing, and running. He also has a passion for fish keeping, which reflects his meticulous nature and love for detail. Mateusz is fluent in Java , Go and Python, and he is constantly exploring new technologies and approaches to improve his craft.
Ticket prices will go up in...
176
Days
:
11
Hours
:
55
Minutes
:
02
Seconds
You missed out!
Venue address
ICE Krakow, ul. Marii Konopnickiej 17
Phone
+48 691 793 877
info@devoxx.pl