Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Berk Mandiracioğlu, a software engineer based in Zurich with a passion for building scalable ML tooling and AI-powered systems for finance and research. I'm focused on delivering impact by shipping reliable tools, frameworks, and pipelines that accelerate model development, deployment, and data analysis for teams across finance and academia.…

Berk Mandiracioglu

Developer, Full Stack Developer, Back-End Developer, +2





I'm Berk Mandiracioğlu, a software engineer based in Zurich with a passion for building scalable ML tooling and AI-powered systems for finance and research. I'm focused on delivering impact by shipping reliable tools, frameworks, and pipelines that accelerate model development, deployment, and data analysis for teams across finance and academia.…

Available to hire

I’m Berk Mandiracioğlu, a software engineer based in Zurich with a passion for building scalable ML tooling and AI-powered systems for finance and research.

I’m focused on delivering impact by shipping reliable tools, frameworks, and pipelines that accelerate model development, deployment, and data analysis for teams across finance and academia.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Work Experience

Software Engineer at UBS

September 1, 2023 - Present

Developed tools to speed up model development and evaluation; built runtime servers to support GenAI agents and ML model deployments for UBS Risk department; created a framework to define LLM Agents capable of executing code and performing data analysis, reducing model development time for statisticians by 50%. Implemented a framework to tune LLM Agent performance and deployed models, cutting deployment time by 2x. Built a VSCode extension for code assistance with over 2,000 daily users. Deployed frameworks to AKS via CI/CD pipelines and managed routes with Ingress.

Software Engineer Intern - Master Thesis at Acodis AG

September 1, 2022 - March 1, 2023

Applied unsupervised clustering methods to document images in production; investigated scalable algorithms to increase throughput, achieving roughly a 5x speedup in document processing; contributed to MlOps for deploying and maintaining models.

Software Engineer Intern at Oracle Labs

September 1, 2021 - January 1, 2022

Developed an anomaly detection platform for SQL logs to identify attacks or system malfunctions; contributed to a benchmarking framework comparing methods from classical statistics to deep learning; reduced the size of the SQL log representation by 60%, increasing throughput by 3x while maintaining accuracy; improved processing of large-scale SQL log datasets by enabling distributed inference.