Hussein Loubani
AI Engineer at CIAD | International Speaker on AI Agents and Computer Vision
Ingénieur IA chez CIAD | Conférencier international sur les agents IA et la vision par ordinateur
Montbéliard, France
Actions
I am an AI Engineer at CIAD, where I build AI systems designed to hold up outside a demo, from real-time perception for autonomous vehicles to agentic workflows wired into real tools and data. The part of AI I care about most is what happens after the prototype, when a model meets messy input, tight deadlines, and users who expect it to work every time.
I speak internationally on AI agents, reliability, evaluation, and designing APIs and tools machines can use correctly. Recent and upcoming stages include Build Stuff in Vilnius, FOST Paris by Apidays, Apidays Munich, Voxxed Days Luxembourg, DutchAICon, IEEE ICRA in Vienna, and JCON OpenBlend in Slovenia. Every talk I give comes from systems I built and broke myself, so the content stays practical, honest, and useful for people who ship.
I also co-organize Google Developer Group communities in France, where I run hands-on AI code labs helping beginners build their first working AI project.
Je suis ingénieur IA chez CIAD, où je conçois des systèmes d'IA pensés pour tenir bon au-delà de la démo, de la perception en temps réel pour les véhicules autonomes jusqu'aux workflows agentiques connectés à de vrais outils et de vraies données. Ce qui m'intéresse le plus dans l'IA, c'est ce qui se passe après le prototype, quand un modèle rencontre des données imparfaites, des délais serrés et des utilisateurs qui attendent qu'il fonctionne à chaque fois.
J'interviens à l'international sur les agents IA, la fiabilité, l'évaluation et la conception d'API et d'outils que les machines savent utiliser correctement. Parmi mes interventions récentes et à venir figurent Build Stuff à Vilnius, FOST Paris par Apidays, Apidays Munich, Voxxed Days Luxembourg, DutchAICon, IEEE ICRA à Vienne et JCON OpenBlend en Slovénie. Chaque présentation s'appuie sur des systèmes que j'ai moi-même construits et cassés, ce qui garde le contenu concret, honnête et utile pour ceux qui livrent en production.
Je co-organise également des communautés Google Developer Group en France, où j'anime des ateliers pratiques d'IA qui aident les débutants à construire leur premier projet d'IA fonctionnel.
Badges
Area of Expertise
Topics
What Self-Driving Cars Taught Me About Reliable AI Agents
Before I built agents, I built real-time perception for autonomous vehicles. A robot does not pause to think. It ingests LiDAR and camera data, runs inference under a deadline, and degrades safely when a sensor drops or a model returns noise. Agents fail the same ways, and most teams rediscover the lessons one outage at a time.
This talk maps reliability engineering from robotics onto agent systems. Deadline budgets and timeouts, so one slow tool does not stall the run. Health checks and fallbacks, the agent version of a dropped sensor. Tracing on every step, so a failure becomes reproducible instead of a mystery. Partial results over crashes, and human checkpoints placed where a wrong action gets expensive.
Every example comes from systems I built and debugged myself, both robot pipelines and agent workflows, including the ones I got wrong first. You leave with a reliability checklist drawn from a field where failure costs more than a bad demo.
Lightweight 3D Maps for Vehicle Localization Without Heavy Point Clouds
Dense point-cloud maps are heavy to store, share, and maintain. My research localizes a vehicle against compact 3D building models instead, using panoramic images and semantic cues to match what the camera sees to a lightweight city model.
In this session I explain the approach without heavy math. I cover how I build the lightweight 3D model, how semantic segmentation links image to map, and how the vehicle tracks its pose against simple building geometry. I show where this beats dense maps on size and upkeep, and where it still struggles.
You leave understanding a lighter path to localization, and the trade-offs against point-cloud methods you may already use. I presented the underlying research as a poster at ICRA 2026 in Vienna, titled Semantic Equirectangular Visual Tracking in Lightweight 3D Building Reconstructions, and this talk adapts it for an automotive open-source audience. All results are my own.
Agentic AI Patterns Developers Can Use Today
Agentic AI is everywhere in 2026, but most examples are either toy demos or huge research systems. This session focuses on practical patterns developers can use right now to add agentic behavior to their applications in a controlled way.
We will walk through simple architectures for LLM-based agents that call tools, work with context, and coordinate steps, while still fitting into normal services, APIs, and UIs. The focus is on what is realistic for a small team, how to keep things observable, and how to avoid agents “going rogue”.
Attendees will learn:
- Core building blocks of an agentic AI system (tools, memory, orchestration)
- Simple patterns to integrate agents into existing backends and workflows
- How to keep agent behavior debuggable and predictable
- Pitfalls to avoid when moving from a chatbot to real agentic features
Build Stuff 2026 Lithuania Sessionize Event Upcoming
AGL All Member Meeting Europe 2026 Sessionize Event
Dutch AI Conference 2026 Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top