Freelancer profile translated to English.
Back to original languageAbout Lilian
AI Architect & Fractional CTO specializing in moving GenAI projects from prototype to production.
I help startups, SMEs, and technical leaderships design, industrialize, and govern truly usable AI systems: AI agents, business workflow automation, RAG / GraphRAG, data pipelines, LLMOps, evaluation, monitoring, security, and cloud infrastructure.
Areas of Expertise:
- AI Agents & Business Automation: Tool orchestration, multi-step workflows, APIs, n8n / Make / Zapier when relevant, human-in-the-loop, integration with existing systems.
- Advanced RAG & Knowledge Systems: Vector databases, GraphRAG, corpus structuring, multi-hop retrieval, reranking, response quality evaluation.
- LLMOps / GenAI MLOps: Production deployment, observability, continuous evaluation, LLM cost management, multi-model routing, Docker / Kubernetes / AWS, robust pipelines.
- AI Governance & Security: Risk assessment, operational compliance, AI Act, traceability, guardrails, protection against uncontrolled usage, monitoring of AI systems in production.
Examples of Projects Completed:
- Large-scale automated customer review analysis and moderation system;
- Data mining and web data structuring pipelines for a major international brand;
- LLM agent architectures and decision scaffolding to automate operational workflows;
- AI maturity assessment, GenAI roadmap, and coaching for executives / technical teams.
French
Native or bilingual
English
Native or bilingual
Can work on-site
Paris (up to 50km)
Experience
- KorbakSpeaker / Trainer LLMOps / GenAI Consulting EngagementsCONSULTING AND AUDITSNovember 2023 - Today (2 years and 7 months)Paris, France
Supporting startups, SMEs, technical leadership, and executive teams in designing, industrializing, and governing generative AI systems in production.
My engagements cover strategic framing, technical architecture, stack selection, production deployment, evaluation, security, governance, and team support.Main Interventions:- AI maturity assessment, use case identification, and GenAI roadmap construction;- Design of AI agent architectures, RAG / GraphRAG, data pipelines, and business automation;- Support for CTOs / technical leadership on build vs. buy decisions, cloud, LLM costs, security, and scalability;- Implementation of LLMOps / GenAI MLOps practices: evaluation, monitoring, observability, multi-model routing, human supervision;- AI governance framing: AI Act, traceability, guardrails, authorized uses, security risks, operational compliance;- Training and acculturation for executives, product teams, and technical teams on the concrete impacts of generative AI.Example Engagements:- Architecture for a large-scale automated customer review moderation system;- Data mining and web data structuring pipelines for a major international brand;- Coaching executives on AI strategy, governance, internal adoption, and product roadmap;- Design of LLM agent systems connected to business tools to automate operational workflows. - MIAKITOFounder & CTO / AI Architect / LLMOps (Agents, RAG & Production Infra)RESEARCHFebruary 2023 - Today (3 years and 4 months)Paris, France
Creation and development of a business dedicated to designing generative AI systems applied to business automation, advanced retrieval, and LLM agent orchestration.
MIAKITO is a product and technical laboratory for developing reusable components around production LLM architectures: agents, RAG / GraphRAG, vector databases, data pipelines, automated moderation, and automation workflows.Main Work:- Design of LLMOps architectures and pipelines of agents connected to business tools;- Experimentation with memory systems, tool orchestration, multi-hop retrieval, and GraphRAG;- Structuring and exploitation of large text corpora;- Implementation of automated classification and moderation systems;- Development of data mining pipelines on customer content, product data, and web data;- Design of evaluation components to measure the quality of responses, robustness, and errors of LLM systems. - Immersive TherapyFounder & CTO, medtech, signal processing & digital productMEDICALNovember 2017 - March 2021 (3 years and 4 months)Rennes, France
Founder and technical director of a medtech startup developing a digital solution for patients suffering from chronic tinnitus.
I managed the entire product and technical cycle: product vision, software architecture, signal processing algorithms, patient application, backend, exploratory clinical trials, and user iterations.Main Responsibilities:- Definition of product strategy and technical roadmap;- Design of software and backend architecture;- Development of algorithms for processing and synthesizing therapeutic audio signals;- Personalized modeling of auditory profiles and tinnitus characteristics;- Design of an inference engine to dynamically adapt sound stimulation protocols;- Management of product experiments, user iterations, and technical trade-offs.This founder/CTO experience now informs my approach to AI projects: building useful, robust systems that are understandable by business teams and capable of moving from prototype to real-world use.
Recommendations
Be the first to recommend Lilian
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Engineering DegreeCentrale Paris2018Diplôme d'ingénieur Majeure Système Embarqués et Réseaux Options Réseaux de Neurones, Algorithmes Génétiques et Intelligence Artificielle
- Preparatory classes for Grandes Écoles (CPGE), MP*CPGE Lycée Pothier2014Classes préparatoires aux grandes écoles (CPGE) MPSI / MP*