Work with Me

From scientific prototype to production system

I support product and R&D teams, scientific project leaders and laboratory directors when a prototype needs to become a reliable, maintainable and operable tool. I work across architecture, software development, data pipelines and the deployment platform, with particular attention to operational simplicity and knowledge transfer.

Areas of Expertise

Modeling & Simulation

Implement numerical models and solve them.

  • Numerical simulations and analysis
  • R&D methodological support
  • Scientific data analysis
Data Engineering

Clean pipelines and analysis your team can use day to day.

  • ETL pipelines
  • Statistical analysis
  • Business dashboards
  • Forecasting & machine learning
Infrastructure & DevOps

A stable platform your team can run in production.

  • Kubernetes deployments
  • CI/CD & GitOps
  • Infrastructure as Code
  • Monitoring & observability

For Laboratories and Scientific Projects

Is your team developing a model, a simulation tool or research code? My focus is scientific software and numerical computing, complemented by support for data, sustainable code and research infrastructure.

Scientific Software & Numerical Computing

Implement and evolve models, automate simulations and optimize computations: libraries and tools suited to your scientific questions.

Data & Reproducibility

Structure data, parameters and processing steps to make analyses and numerical experiments traceable and reproducible.

Sustainable Research Code

Simplify and refactor code, add tests and documentation, then share the practices the team needs to take it over.

Research Infrastructure

Set up computing environments, workflows and parallel execution to support simulations and data processing.

Concrete Examples

LOCIE — pySIP

2023
Context
A Python library for system identification and stochastic modeling to rewrite and simplify.
Contribution
Full rewrite and simplification of pySIP, delegation of Bayesian inference to PyMC3 and performance optimization with Numba.
Delivered Capability
A reworked library to identify models, use PyMC3’s Bayesian inference and run the Numba-accelerated KalmanQR filter.
pySIP — source and migration notes

Industrial Client — Scientific Computing

2024
Context
Solve partial differential equations on unstructured domains.
Contribution
Automated PDE solving using Trevor Hines’s RBF library as the reference solver.
Delivered Capability
An automated solving workflow on unstructured domains, backed by this reference solver.
Trevor Hines’s RBF — third-party library used

Industrial Client — Energy Sector

2025
Context
Explore the parameters of a specific EnergyPlus configuration template.
Contribution
Automated parametric studies through design of experiments and parallel simulations, using my Python wrapper to launch EnergyPlus programmatically.
Delivered Capability
Run simulations in parallel from the experimental design applied to the EnergyPlus template.
My EnergyPlus wrapper — public source

Ways of Working

Strategic Advisory

We clarify the need, then choose a realistic technical approach.

Embedded Team Support

Join your product, R&D or research team to contribute to code, reviews and knowledge transfer.

Defined-Scope Mission

Agree on scope, deliverables and validation criteria, then take ownership of implementation and handover to the team.

Let's Talk About Your Project

Tell me about your product, R&D or scientific need, the current state of the code and your preferred format: a mission with defined scope and deliverables, or embedded support within your team.

contact@celliern.io