Home

My Profile

HELLO EVERYBODY, I AM

Jean Bellot, PhD

CTO, Lead Datascientist, AI researcher

I’m a Data Scientist and AI Advisor with a Master’s degree in Artificial Intelligence and a PhD in Computational Neuroscience. For 8 years, I’ve served as Lead Data Scientist and CTO in a start-up, leading teams to design and deploy intelligent systems that create real impact — from strategy and architecture to implementation and product delivery.

Since Jan 2024, as a freelance consultant, I help companies make the most of AI — by identifying where it can bring real value, designing intelligent tools adapted to their unique use cases, and helping them to build systems that are both technically robust and human-centered.

AI should enhance human experience, focusing on tedious work allowing us to focus on creativity. I believe the true power of AI lies in augmenting human potential, simplifying complexity, and creating tools that empower people to focus on what truly matters. Every project I take on is guided by that principle — using AI not as an end in itself, but as a meaningful force for better decision-making, creativity, and human connection.

19th November, 1988
+33 (0)695632586
jeanbellot@gmail.com
Paris, France

Services

AI & Data Strategy

Identify where AI creates real impact, design actionable roadmaps, and define metrics for ethical, sustainable implementation.

Data Science & Machine Learning

From data exploration to deployment: predictive models, forecasting, and data-driven decision tools.

Agentic & RAG Systems

Build custom Agentic workflow and Retrieval-Augmented Generation solutions adapted to your domain.

AI Product Development

Technical leadership and hands-on support for designing, prototyping, and scaling intelligent products that enhance human experience.

Advisory & Mentorship

Strategic guidance to help teams and leaders understand, adopt, and grow with AI responsibly.

Experience

Work Profile

Since Jan 2024 Free Lance as a Technology and AI advisor @ Lazy Smart (not an official brand yet) Help client understand how AI and digital technology can help them in their business. Image generation, LLM, RAG, Visual Recognition, Forecasting...
Since Sept 2024 CTO @ A Lab Curated By
Jan 2020 - Jan 2024 Chief Technology officer and Lead Datascientist @ Heuritech Manage prioritization of all tech teams at Heuritech (R&D, Data Science, Infrastructure, Product, Data Engineering, Internal Tools) and Lead Datascience
Jan 2021 – Jan 2024, PhD Co-director of Etienne David's PhD Time series forecasting models applied on large datasets with inclusion of external signals in collaboration Sylvain Le Corff, Telecom Paris
2018 – 2020, Technical Lead Product & Data science @ Heuritech Manage product team, build and manage the data science team, people and roadmap
2016 - 2018, Datascientist @ Heuritech Member of the R&D team working on visual recognition and Data Science
2011 – 2015, PhD Student @ Institut des Systèmes Intelligent et de Robotique Université Pierre et Marie Curie – Paris VI Phd Thesis in computationnal neuroscience
Computationnal Modeling of the role of dopamine in the corticostriatal loops in action selection
2012 – 2014, Teaching Assistant @ Politech Paris Université Pierre et Marie Curie - Paris VI, in Software Engineering

Education

2011 – 2015, PhD Student @ Institut des Systèmes Intelligent et de Robotique Université Pierre et Marie Curie – Paris VI Phd Thesis in computationnal neuroscience
Computationnal Modeling of the role of dopamine in the corticostriatal loops in action selection
2009 - 2011 Master Degree in computer science Universite Pierre et Marie Curie - Paris VI Artificial Intelligence and Decision
Research Specialization: Intelligent Agent, Learning & Decision
2006 - 2009 Licence Mathematics & Computer Science Universite Pierre et Marie Curie – Paris VI Specialty Statistics & Artificial Intelligence

Publication

(Currently in review), Bellot, J., Sigaud, O., Roesch, M.R., Shoenbaum, G., Girard, B., & Khamassi, M. Dopamine neurons phasic activity does not encode the reward prediction error that behavioral adaptation would predict Neurons, Bheavior, Data analysis and Theory
2023, David, E., Bellot, J., & Corff, S. L. Hermes: Hybrid error-corrector model with inclusion of external signals for nonstationary fashion time series Transactions on Machine Learning Research
2023, David, E., Bellot, J., Corff, S. L., & Lehéricy, L. Identifiability of discrete input–output hidden markov models with external signals Statistics and Computing
2015, Bellot, J. Computationnal modeling of the role of dopamine in the corticostriatal loops in action selection. Doctoral dissertation, Université Pierre et Marie Curie, Paris VI
2014 Bellot, J., Liénard, J., Khamassi, M., & Girard, B. Investigation of the effect of different dopaminergic levels on the basal ganglia action selection properties using a biologically constrained computational model SfN neuroscience meeting. Washington. poster 555.23/TT42
2013 Bellot, J., Liénard, J., Khamassi, M., & Girard, B. Which temporal difference learning algorithm best reproduces dopamine activity in multi-choice task? Third Symposium on Biology of Decision - Making (SBDM 2013). Paris. Poster 5
2012a Bellot, J., Sigaud, O., & Khamassi, M. Dopamine neurons activity in a multi-choice task: Reward prediction error or value function? Proceedings of the French Computational Neuroscience NeuroComp/KEOpS1́2 workshop
2012b Bellot, J., Sigaud, O., & Khamassi, M. Which temporal difference learning algorithms best reproduce dopamine activity in a multi-choice task?
From Animals to Animats: Proceedings of the 12th International Conference on Adaptive Behaviour (SAB 2012) Obtained the best paper award of the conference
2012 Bellot, J., Sigaud, O., Roesch, M. R., Schoenbaum, G., Girard, B., & Khamassi, M. Dopamine neurons activity in a multi-choice task: Reward prediction error or value function? Proceedings of the French Computational Neuroscience NeuroComp/KEOpS1́2 workshop. Pages 1-7. Bordeaux, France.

Teaching

Miscellaneous Experience

2012 – 2014 Introduction à la programmation orienté objets en Java Politech Paris, ELI3 Teaching assistant (2*10h) and E2I4 Teaching assistant (32h).
2013 Introduction à la programmation en Python, GM3 GM3 teaching assistant 20h
2014 Introduction à la compilation, Tex/Lyx Teaching assistant 20h

Projects

2011 Comparison between dopamine activity and reward prediction error in a multi-choicetask Institut des Systèmes Intelligents et de Robotique Internship M2 in Computational Neuroscience
2010 Supervised classification of textual documents with boosting Laboratoire d’informatique de Paris VI - Lip6 Internship M1 in Artificial Intelligence

Skills

Languages French, native
English, fluent with a lovely french accent
German, Ich habe Deutsch gelernt aber alles vergessen
Coding Python, bash, pandas, scikitlearn, openCV, vim, tmux, airflow, prefect, git, pytorch, huggingface, LangGraph, LangChain...
Deployment docker, docker-compose, gitlab (CI/CD), k8s (basic level), PostgreSQL
Orchestration Prefect 1 & 2, Airflow
Project Management JIRA, Trello, Monday, Notion, ProductPlan