Miklos Szegedi is a software engineering specialist and founder with nine years of formal experience and a two-decade technical pedigree spanning embedded drivers to large-scale distributed systems. He builds tech assets aimed at institutional investment and retirement funds while also running stealth AI ventures and open-source reference projects for AI code generation. A pragmatic full-stack engineer, he has shipped storage and compute features at Amazon Redshift, contributed backend fixes to Apache Hadoop/YARN, and led autoscaling work in Go at Cloudera. Currently focused on AI/LLM training at SpaceX and co-founding an AI-native transaction verification startup, he blends low-level systems knowledge (chip testing, codecs, firmware) with cloud, data streaming, and ML stacks. Beyond engineering, he serves as a substitute teacher and holds an MBA, signaling an uncommon mix of technical depth, operational leadership, and community engagement. He publishes under a pen name and funds open-source R&D through contract work, turning contracting proceeds into reusable AI and RAG reference solutions.
9 years of coding experience
24 years of employment as a software developer
German as a Second Language, German as a Second Language at Österreichische Institut
Master of Business Administration - MBA, Business Administration and Management, General & Finance, Master of Business Administration - MBA, Business Administration and Management, General & Finance at Louisiana State University
Ballroom Dancing, Ballroom Dancing at New York Dance Association, Budapest, Hungary
M.Sc., Information Technology - Computer Science, M.Sc., Information Technology - Computer Science at Budapest University of Technology and Economics
Certificate, Digital Marketing, Certificate, Digital Marketing at Cornell University
Contributions:47 commits, 14 PRs, 73 comments in 7 months
Contributions summary:Miklos contributed to the Apache Hadoop project by addressing memory limit issues and improving logging in the YARN framework. They modified the `ContainersMonitorImpl` class to provide more detailed error messages when memory limits are exceeded. Furthermore, they improved the `ContainerLogAppender` and `ProcfsBasedProcessTree` components, enhancing logging and process tree functionalities within the YARN ecosystem. They also addressed a bug within the FairScheduler framework.
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.
Request Free Trial
Miklos Szegedi - Software Engineering Specialist at SpaceX