Thierno Diop
Freelance ML Engineer at Karat
Dakar, Dakar Region, Senegal
Summary
🤩
Rockstar🎓
Top SchoolThierno Diop is a versatile ML engineer and entrepreneur with 11 years of experience bridging NLP/deep learning research, production ML systems and full-stack development. As a freelance ML engineer and former lead data scientist, he has shipped phoneme detection pipelines, RAG-based chatbots, LLM-driven vulnerability analysis for Ethereum, and multi-agent systems—combining hands-on model training with scalable data and indexing pipelines. He mentors data science students, co-founded Senegal’s GalsenAI community, and serves as a Google Developer Expert and Coursera instructor, signaling strong technical leadership and local ecosystem impact. An active open-source contributor, he has improved OCR recognition/training code in keras-ocr and helped refine TensorFlow docs, demonstrating attention to both model internals and developer experience. Notably, his background in mobile/web development enables him to move projects from prototype to production across the stack.
11 years of coding experience
7 years of employment as a software developer
Diplôme d'ingénieur Programmation / développeur informatique général, Diplôme d'ingénieur Programmation / développeur informatique général at Ecole supérieur polytechnique de DAKAR(ESP)
Stackoverflow
159reputation
59kreached
9answers
8questions
Github Skills (19)
learn-ruby-on-rails10
python10
ocr10
user-manual10
machine-learning10
basics10
keras10
tensorflow10
ocra10
computer-vision10
documentation10
nlp6
sentiment-analysis6
php6
symfony6
Programming languages (10)
Github contributions (5)
faustomorales/keras-ocr
Jan 2020 - Jan 2020
A packaged and flexible version of the CRAFT text detector and Keras CRNN recognition model.
Role in this project:
ML Engineer
ML EngineerContributions:7 commits, 5 PRs, 12 comments in 11 days
Contributions summary:Thierno primarily contributed to the `keras-ocr` repository by making changes related to the recognition model and training scripts. They modified the code to use `-1` as a placeholder for empty tokens and updated the docstrings of the `recognize` method, indicating work on model functionality. The user also worked on explicitly defining and utilizing `max_string_length` within the training scripts and using the recognizer to dynamically determine this value.
deep-learningrecognitionocrdetectorcraft
tensorflow/docs
Mar 2019 - Aug 2019
TensorFlow documentation
Role in this project:Technical Writer
Technical WriterContributions:18 commits, 12 PRs, 8 comments in 4 months
Contributions summary:Thierno's contributions primarily consist of fixing typos and minor textual errors within the TensorFlow documentation. Their edits are focused on the migration guide, tutorials, and introductory materials, ensuring accuracy and clarity of the documentation. The changes involve correcting spelling mistakes, refining sentence structures, and updating outdated links and code snippets, thereby improving the quality and usability of the documentation. This includes updating colab links and resolving clarity issues.
deep-learningmachine-learningdeep-neural-networkstensorflow-tutorialstensorflow
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.