Experience

Professional experience

The same thread at every step: complex systems, measurable results.

1. August 2025 - April 2026

   Founding AI Engineer & Architect

   Arkim AI, pre-seed startup · Freelance mission, remote (Los Angeles)

   Alongside the founders: from product vision to architecture, through to shipping the AI system to production.

   • Production multimodal ingestion pipeline on Databricks: OCR, image-to-text, semantic chunking and embedding at scale.
   • Two-level multi-agent system with LangGraph: Plan-and-Execute orchestration of ReAct sub-agents, human-in-the-loop checkpoints, SSE streaming via FastAPI and prompt caching on AWS Bedrock.
   • RAG evaluation framework (Ragas + LangSmith) across 5 pipeline candidates: the in-house pipeline outperformed the commercial baseline with a Precision@k of 0.82 vs 0.68 and a Recall@k of 0.70 vs 0.56.
   • Observability and feedback loops put in place to enable data-driven product iteration.

   Case studiesMulti-agent architectureMultimodal RAG and evaluation

   Marius was a founding engineer at Arkim, where he designed and deployed core parts of our AI platform. Including our multimodal RAG pipeline and agentic workflows built on LangGraph. Working across time zones with minimal oversight, he consistently took ambiguous problems and came back with shipped, production-ready solutions. I'd bring him in again without hesitation.

   See on LinkedIn

   Working alongside Marius at Arkim has been at absolute pleasure and I have no doubt that he will continue to excel in any technical role that he chooses to pursue. Like myself, he's a guy who came from heavily regulated industry (aerospace, in his case) and developed a passion for building reasoning systems with AI tools that solve real-world problems. As CTO, I worked very closely with Marius developing agent orchestration tools and context management systems where I witnessed the following first-hand: 1. he is brilliant 2. he produced the best architecture diagrams and test reports I've seen in my career 3. he's very quick to learn, innovate, and adapt I sincerely hope to work together again in future projects and would highly recommend Marius to any team in need of AI expertise.

   See on LinkedIn

2. October 2023 - August 2025

   Independent AI Engineering

   Independent · Paris, France

   A deliberate transition from industrial engineering to AI engineering, run as an R&D program: systems built end to end, plus a Data Analytics program at the Sorbonne (2025).

   • Systems delivered from exploration to production: industrialized credit scoring (MLOps), a real-time streaming pipeline on Kafka, applied deep learning (vision, NLP, time series), RAG solutions and open-source model deployment.
   • Every project is documented: architecture, method and results.

   See the projects in the portfolio

3. October 2022 - September 2023

   Production Lead, maintenance documentation

   Docaret, for Dassault Aviation · Saint-Cloud, France

   Co-managed the outsourced production of Rafale maintenance documentation: three teams in France and internationally, technical writing standards, expert reviews for technical consolidation, XML compliance and version control.

   • Workflows restructured: backlog absorbed, then contractor production capacity increased by up to 50%.
   • Business Intelligence processes and tools designed for real-time production tracking, internally and externally.

4. October 2019 - September 2022

   Industrialization Engineer (apprenticeship)

   Safran Electrical & Power · Niort, France and Pitstone, United Kingdom

   Industrialization of automated assembly lines: hardened an automated cell in production, designed an assembly line for the ENGINeUS electric motors, ran automation and weight-reduction studies.

   • Lean Six Sigma Green Belt project on an automated cell: OEE from 40% to 87%, scrap from 8% to 2%, Cp/Cpk capability from below 1 to above 2, 28 quality nonconformance reports brought to zero, €19,419 in annual savings. Power BI put in place for business intelligence on the production means.
   • Designed a semi-automatic assembly line for the ENGINeUS motors: proof of concept de-risking the most critical assembly operation, 10 line configurations studied, specifications carried through to integrator consultation.
   • Led the process PFMEA for the ENGINeUS motor assembly at the Pitstone site: 176 risks scored in a week and a half with 8 disciplines, control plan included.
   • Relocated an assembly from Pitstone to France in the wake of Covid-19: full re-industrialization of the power unit assembly and test process.
   • R&D busbar weight-reduction study: up to 30% lighter, demonstrated by calculation and validated in thermal testing, at near-zero manufacturing overcost.
   • Cobot automation study for a manual process: solution sized at €79,530 in annual gains, payback in 1.94 years.

5. September 2017 - September 2019

   Methods Technician (apprenticeship)

   Safran Electrical & Power · Niort, France

   First years in methods engineering, at the heart of automating a production sector: time and cost analysis, flows, production documentation and standards.

   • Re-laid-out a production sector and integrated three automated production means, including a robotic cell: layout, flows, PFMEA and machine documentation. The same cell I would later harden as a Green Belt project.
   • Flow reorganization and 5S standards: kanban, dedicated preparation station, reduction of non-value-adding movements and FOD risk.