Mark Sze
Senior Software Engineer at AG2
Sydney, New South Wales, Australia
Summary
🤩
Rockstar🎓
Top SchoolMark Sze is a Senior Software Engineer based in Sydney who focuses on applying agentic AI to real-world business problems and aligning multi-agent LLM workflows with product needs. With five years in hands-on AI engineering and a long technology career that includes founding a startup and regional leadership roles, he blends technical delivery with commercial sensibility. He is an active maintainer on Microsoft Research’s AutoGen, contributing backend code and practical documentation that expanded support for non‑OpenAI models like LM Studio and Mistral. At AG2 he supports an open AgentOS ecosystem, helping developers and businesses integrate agentic capabilities into production. Known for turning research-grade tooling into usable developer experiences, he pairs deep systems experience—from legacy mainframes to modern LLM stacks—with a pragmatic focus on adoption.
5 years of coding experience
9 years of employment as a software developer
Computing Science, Computing Science, Computing Science, Computing Science at University of Technology Sydney
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (9)
autogen10
agent10
ai-agent10
python10
documentation10
llm10
pytest9
agentic9
testing9
Programming languages (5)
Github contributions (5)
microsoft/autogen
Jan 2024 - Nov 2024
A programming framework for agentic AI
Role in this project:
Backend Developer & Technical Writer
Backend Developer & Technical WriterContributions:160 reviews, 32 PRs, 177 pushes in 9 months
Contributions summary:Mark primarily focused on enhancing the documentation and adding features related to non-OpenAI models within the AutoGen framework. Their contributions included creating documentation sections, adding examples, and providing guidance on using different models like LM Studio and Mistral. The user also implemented features like the ability to specify the 'role' field for select speaker messages in Group Chats. They made code changes in the groupchat, test, and documentation files.
agentic-aichatgptllm-agentllm-frameworkagentic
marklysze/autogenlocalllm
Feb 2024 - Apr 2024
AutoGen tweaks for better Local/Alt LLM support
Contributions:39 pushes, 19 branches in 1 month
autogenllm
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.