Guillaume Jeffroy

Spinoff from LAAS-CNRS, Senscell Biosystems commercializes a nanoelectrode biochip technology capable of simultaneously recording the electrical activity of cells on dozens of channels. I joined the company at its inception to design and develop the data exploitation software from scratch.

The project progresses continuously through successive mini-POCs: each feature is first prototyped and validated in isolation before being integrated into the main tool. This approach allows for the delivery of usable software at each stage, reduces integration risks, and enables adjustments as scientific needs become clearer.

The software now covers the entire exploitation workflow and is enhanced with ongoing experiments:

- **Navigation and playback**: Freely browse any time window of a long acquisition, replay channel by channel or across the entire electrode grid, without waiting for the entire file to load.

- **Grid visualization**: A heatmap view mirroring the physical layout of the biochip to quickly identify active areas, silent electrodes, or artifacts.

- **Signal cleaning**: Automatic preprocessing adapted to the specific noise profile of each experiment, with manual adjustment options.

- **Analysis and export**: Summary statistics per channel and per session, exportable to Excel or structured formats for reporting.

- **Traceability**: Each analysis is logged with its parameters and results, allowing it to be retrieved, re-run on new data, or compared across experiments.

Heavy computations and visualization rendering run in the background to ensure a fluid and responsive interface, even with large acquisitions—a must-have for the tool to be truly usable in daily experimental contexts.

Design of a conversational agent to generate and analyze temporal signals. The goal is to make signal exploration and processing accessible: one can describe what they want to do in natural language, and the agent guides the user then launches the appropriate tools (simulation, filtering, event detection, time-frequency analysis).

The project strongly emphasizes performance and interactivity: ClickHouse is used to efficiently store and query large time series, Julia executes heavy computations (signal simulation, multi-band DWT, filtering pipelines, spike detection), and uPlot ensures fast and stable web visualization even with large volumes. The whole system is orchestrated via a FastAPI backend with real-time streaming (SSE) and a chat/action interface that allows chaining analyses while maintaining a clear user experience.

Design and development of a platform for mass processing of HR documents (CVs, job descriptions, training records, job offers, etc.) in heterogeneous formats (PDF, DOCX, Excel). The system automatically extracts, structures, and harmonizes key information using multi-model AI pipelines (LLM, Sentence Transformers) and parallelized workflows.

The architecture relies on decoupled microservices (FastAPI / Prefect), deployed in containers on French cloud infrastructure (Scaleway), with S3 object storage and a real-time monitoring dashboard. A checkpoint mechanism ensures automatic error recovery, enabling a success rate close to 100% on over 10,000 processed documents.

GDPR compliance is natively ensured by an anonymization and PII detection pipeline integrated into the processing flow.

Python · FastAPI · Prefect · LLM · Sentence Transformers · Docker · Scaleway · React