Anca Dumitrache
Machine Learning Scientist II at Booking.com
Amsterdam, North Holland, Netherlands
Summary
👤
Senior🎓
Top SchoolAnca Dumitrache is a Machine Learning Scientist with 14 years of experience blending NLP, recommender systems, HCI and crowdsourcing into production ML solutions. She holds a PhD in Artificial Intelligence from VU Amsterdam where her research showed how inter-annotator disagreement can improve training data quality—insight she later applied at IBM Watson and Google Research internships. At product companies like Albert Heijn and Booking.com she built and deployed PySpark/TensorFlow pipelines for personalized ranking, product embeddings and uplift models, while mentoring juniors and running internal NLP initiatives. Her background spans academic rigour and hands-on engineering across cloud platforms (Azure, AWS) and languages (Python, R, Java), reflecting a rare ability to move ideas from crowd-collected signals to scalable systems. Anca’s work often combines semantic understanding with user-centric evaluation, revealing useful patterns in noisy human data that standard approaches miss.
14 years of coding experience
11 years of employment as a software developer
Doctor of Philosophy - PhD, Artificial Intelligence, Doctor of Philosophy - PhD, Artificial Intelligence at Vrije Universiteit Amsterdam (VU Amsterdam)BSc, Computer Science, BSc, Computer Science at Jacobs University BremenColegiul Național "Liviu Rebreanu" Bistrița, Romania
English, Romanian, French, German, Dutch
Stackoverflow
11reputation
486reached
2answers
2questions
Github Skills (30)
relation-extraction10
python10
human-computer-interaction10
annotations10
ambiguity-resolver10
machine-learning10
annotation-processing10
deep-learning10
tensorflow10
data-science9
crowdsourcing9
artificial-intelligence9
truth9
ml9
natural-language-processing9
Programming languages (7)
Github contributions (5)
CrowdTruth/Medical-Relation-Extraction
Sep 2015 - Mar 2019
Crowdsourced ground truth for medical relation extraction.
Contributions:5 releases, 52 commits, 46 pushes in 3 years 6 months
relation-extraction
CrowdTruth/CrowdTruth-core
Apr 2017 - May 2019
CrowdTruth framework for crowdsourcing ground truth for training & evaluation of AI systems
Contributions:5 releases, 106 commits, 86 pushes in 2 years 1 month
aicrowdsourcinginter-annotator-agreementhuman-computer-interactionannotations
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.