About Philippe
Let's work together to bring your project to reality faster and cleaner!
Whether it's about
- prototyping a new AI solution
- improving your backend
- accelerating time to market
- scaling an existing codebase
- refactoring technical debt
- or taking on a complete project autonomously
I am here to do it efficiently and reliably.
Drawing from my experiences in the fast-moving AI field and fast-growing startups (Preligens) I am confident in my ability to quickly become a valuable asset to your team.
As an AI teacher, I have been working with international colleagues and students. I love adapting to knew environments and cultures. I speak French and English fluently.
Skilled in design and development of
- cutting-edge AI systems (training and inference)
- backend
- API
- Database
- Production environments
Technical Stack
- Python
- Typescript
- PostgreSQL, Neo4j
- GraphQL
- FastAPI, Flask, Django
- Tensorflow, Pytorch
- Scikit-learn
- Numpy, pandas
- Docker, docker buildx
- Git, github, gitlab
- Agile
- AWS EC2, AWS S3
- Google Cloud Compute
- Reading the documentation
And much more libraries and tools
French
Native or bilingual
English
Fluent
Can work on-site
Paris (up to 30km), Marseille (up to 30km), Lyon (up to 30km), Bordeaux (up to 30km), Toulouse (up to 30km)
Experience
- CirclBackend / Data Engineer - LLM SpecialistE-COMMERCEApril 2024 - April 2024 (1 month)Paris, FranceEnhancing Data Cleaning and Integration with LLMIn collaboration with the CTO of Circl, I spearheaded the development of a transformative feature: the complete automation of complex scraped data cleaning and its integration via a ETL process.The need to use a LLM arose from the complexity of the data, which previously required manual handling. LLMs are capable of understanding the context of the data, its nuances and to ignore the noise without much humain intervention. The perfect tool for this job.Key achievements include:
- Modelled a system to categorize the data, enhancing the effectiveness of LLM model prompts.
- Created a provider-agnostic tool using Langchain to streamline the use of various LLMs including OpenAI GPT-4 and MistralAI models.
- Deployed the open-source Mixtral 8x7B model on Google VertexAI.
- Developed new FastAPI endpoints to expose the categorization system.
- Utilized BigQuery for the efficient storage of validated LLM outputs, ensuring data integrity.
- Integrated the new cleaning process into the existing Google Workflow ETL, allowing complete automation of the data ingestion pipeline
- Propagated the new data to the databases via DBT
- Quickly built a Python CLI tool using Typer to manage data and facilitate ongoing experimentation with new data sets and models.
This project leveraged the power of LLMs to automate the data cleaning process, thereby reducing the time and effort required to handle the data. And enabled continuous innovation in handling and utilizing large language models. - DiagorienteBackend Engineer - Data EngineerPUBLIC SECTOROctober 2023 - March 2024 (5 months)With the team at Diagoriente (beta gouv) with have- integrated a new ETL tool (dagster - python)- added new pipelines and data sources- deployed a new ArangoDB "Graph" database- developped new APIs (FastAPI + SQLAlchemy)- refactored a lot of legacy code- established a modular monorepo- exposed these new tools via CLIs and web-based Graphical User Interfaces (streamlit, react)- dockerised all of our applications- optimized SQL and AQL (arango) queries
- LaliloBackend Engineer - ML EngineerEDUCATION AND E-LEARNINGMay 2023 - November 2023 (6 months)Paris, FranceI assist Lalilo and its engineers in deploying Machine Learning / Deep learning models into production. In more details, I collaborate with my teammates to do the following:- Design the pipeline to ship deep learning models to production (Pytorch - Speechbrain)- Run our production systems on k8s- Design and implementation of our Flask API- Write our internal facing documentation- Develop automated testing of our API- Refactor the existing code base- Improve processesTools- python- typescript- flask, FastAPI- sqalchemy, alembic- pytorch- postgreSQL- docker- kubernetes- aws- datadog- test driven development (TDD)- agile
Reviews
Recommendations
Be the first to recommend Philippe
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Architecte du numerique422018Project based learning. C / Python / Backend programming / shell
- Prepa PCSI / PSILycee Jean-Baptiste Say2016"Physique, Chimie, Sciences de l'ingenieur" / "Physics Chemistry and Engineering Sciences" • Emphasis on advanced mathematics, including algebra, analysis, and geometry • Study of classical and modern physics, including mechanics, electromagnetism, and optics • Study of chemistry: basics of inorganic and organic chemistry, as well as thermodynamics and kinetics • Electronics: basics of electronic circuits and their applications in physics and engineering • Engineering: kinematics, dynamics, and thermodynamics. As well as, materials science and robotics (including control systems, sensors, and actuators) • Introduction to computer science and programming Development of problem-solving and critical thinking skills
Certifications
- Machine Learning - Stanford UniversityCoursera2017
- Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and OptimizationCoursera2017