Florent Vaucher

Projects

{

"name": "Florent",

"occupation": "Machine Learning Engineer",

"interests": ["Coding", "Science"],

}

My Projects

Brain Diseases Diagnostic

CERN LHC Chatbot

Portfolio

Train 3 different models to diagnose Alzheimer's, brain tumors, and brain strokes from CT scans.
Implemented a backend API to select the model and send back diagnosis predictions with certainty level.
Developed a frontend user interface to upload pictures and obtain results.
Handles multiple files, analyses them asynchronously and sends back results as soon as they are ready.
Achieved state-of-the-art performance on 2 out of the three tasks.
Applied advanced computer vision training techniques, such as stratified-k-fold and data augmentation.

Custom-built chatbot tailored for inquiries related to the Large Hadron Collider (LHC)
Created a pipeline to extract information from unstructured data to a base knowledge database
Ensured accuracy of responses by efficiently fetching relevant information
Delivered swift responses through an intuitive and user-friendly interface
Ready for deployment on cloud
Designed to streamline inquiries and provide reliable information about the LHC

Employs advanced speech recognition algorithms for accurate language detection
Supports real-time translation from microphone input or audio files
Generates synthetic audio and text translations in the target language with minimal latency
Accessible via a user-friendly web interface for seamless integration into various applications and workflows
Offers efficient and reliable speech translation services for multilingual communication, language learning, and accessibility purposes
Designed to meet diverse needs with customizable features and flexible deployment options

Voice Translator

See on github

My Work

Software Developer Intern,

MeteoSwiss

LLM Engineer Intern, Innovatim

Experience

Spearheaded initiatives to modernize outdated services and contribute to digital transformation efforts
I independently developed a new project, now essential to MeteoSwiss's arsenal for safeguarding the Swiss population, particularly in predicting and monitoring heatwaves
Pioneered the MeteoSwiss HealthTools Project and led the integration of AWS S3 for forecast
Implemented best practices of software development to ensure high-quality code and efficient processes throughout tenure.
Boosted test coverage of critical services by 80%, ensuring robustness and reliability of software systems.

Developed innovative AI solutions by integrating advanced language models (LLMs), utilizing machine learning frameworks for process automation
Provided AI expertise for decision-making and conducted extensive testing, prototyping, and research to identify the most suitable technologies
Implemented state-of-the-art models on GPU infrastructure, and deployed solutions in the Infomaniak cloud environment.

About Me

I am a dedicated and passionate software engineer with a strong background in machine learning, artificial intelligence, and full-stack development. Throughout my career, I have continually sought opportunities to innovate and contribute to meaningful projects that push the boundaries of technology.

Portfolio

My journey in software development began during my internship at MeteoSwiss in Geneva, where I played a pivotal role in modernizing outdated services and spearheading digital transformation initiatives. I led the development of a new project that is now integral to MeteoSwiss's efforts in predicting and monitoring heatwaves, effectively safeguarding the Swiss population.

My approach to software development is grounded in a commitment to excellence, continuous learning, and a strong emphasis on delivering impactful results. I am adept at identifying and implementing the most suitable technologies through rigorous testing, prototyping, and research, ensuring that each project is equipped with the tools necessary for success.