[ Tech stack ]
MLflow
The open-source MLOps platform that tracks, versions and ships your models.
MLflow logs every experiment (parameters, metrics, artifacts), centralizes models in a versioned registry with aliases and tags, then packages them for serving, with evaluation and tracing for LLM applications. Open source, framework-agnostic, managed on Databricks and available on the major clouds, it's our choice to get ML out of the notebook and make it reproducible and auditable.
[ Why MLflow at Dexon ]
What this technology does well, and why we use it.
Typical usage: Experiment tracking, model registry, ML and LLM production rollout.
- 01
Tracking: parameters, metrics and artifacts of every run, side by side.
- 02
Model Registry: versions, aliases and tags, controlled promotion to prod.
- 03
Standard packaging served as a REST API, Docker image or cloud endpoint.
- 04
Evaluation and LLM tracing: prompts, agents and RAG made observable.
[ Complementary technologies ]
The building blocks we often mobilise alongside.
A stack rarely exists alone. Here are the technologies Dexon most often pairs with this one, through pipeline habits, usage similarity or internal mastery. Click on a building block to see its scope.
[ Reassurance ]
- 0+
- custom projects delivered
- 30
- engineers, designers, project managers
- 80 %
- from top French schools
- 24 h
- average reply time
[ They trust us ]
More than 100 French and European companies trust us
[ Press ]
They talk about us.
Nationwide coverage by BFM Business, Le Figaro, Challenges, La Tribune and CNews. An outside reading of our work and our innovations.
