Giulio Starace
Founder at Paradigma
Rome, Lazio, Italy
Summary
🤩
Rockstar🎓
Top SchoolGiulio Starace is a founder and research engineer with eight years of experience building ML-driven evaluation pipelines and production-ready tooling from Rome. He has shipped multiple high-profile evaluation suites at OpenAI (PaperBench, MLE-bench, SWE-bench Verified) and contributed to the open-source openai/evals framework, adding multi-provider solvers and analysis tooling. His background spans academic research in cross-lingual adaptation of LLMs, astrophysical forecasting pipelines at ESO, and real-time media ML, giving him a strong mix of research rigor and engineering pragmatism. Giulio is comfortable across back-end, MLOps and data pipelines, and has a knack for integrating diverse APIs and instrumenting model evaluation at scale. He founded Paradigma to apply this expertise commercially, translating complex evaluation and deployment problems into reproducible systems. Multilingual and internationally educated, he pairs deep technical skills with a global, interdisciplinary perspective.
8 years of coding experience
4 years of employment as a software developer
High School, International Baccalaureate Diploma Program, High School, International Baccalaureate Diploma Program at Marymount International School RomeMaster of Science - MS, Artificial Intelligence, Master of Science - MS, Artificial Intelligence at University of AmsterdamBachelor of Science, Natural Sciences (Physics & Computer Science), Bachelor of Science, Natural Sciences (Physics & Computer Science) at Durham UniversityHigh School, High School at American Community School of Abu DhabiAmerican Embassy School, New DelhiIstituto Villa Flaminia
English, Italian, Spanish
Stackoverflow
380reputation
42kreached
6answers
5questions
Github Skills (16)
apim10
openai-api10
evaluation-framework10
api10
ant10
back-end-development10
python10
pandas6
unit-testing6
tensorboard6
nodejs6
machine-learning6
time-series6
tensorflow6
numpy6
Programming languages (20)
Github contributions (5)
openai/evals
Nov 2023 - Apr 2024
Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.
Role in this project:
Back-end Developer / MLOps Engineer
Back-end Developer / MLOps EngineerContributions:3 reviews, 10 PRs, 5 comments in 5 months
Contributions summary:Giulio contributed to the `evals` framework, specifically by modifying existing code and adding new features. Their work includes dropping datasets related to steganography and fixing an issue in `OpenAIAssistantsSolver`, indicating a focus on model evaluation and related infrastructure. They also implemented an AnthropicSolver, demonstrating expertise in integrating different API providers and managing their specific requirements. Further contributions included a TogetherSolver and code for making plots and token counts, highlighting the user's involvement in setting up the framework for evaluation and analysis.
llm
thesofakillers/cv
Jun 2019 - Dec 2025
Contributions:44 pushes, 5 branches in 6 years 7 months
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.