Spatial AI Architect — Program Syllabus

Reference syllabus for the flagship Spatial AI Architect program delivered by the 3D Geodata Academy. It defines the learning objectives, audience, technical requirements, the module-by-module program, the assessment scheme and the legal terms of purchase.

"Become the architect of the digital world. Master the full 3D data lifecycle, from raw acquisition to autonomous AI agents and intelligent Scan-to-BIM."

1. Program Overview

Dimension	Details
Format	Six modules, 100% online, asynchronous (FOAD). Optional synchronous mentoring sessions.
Price	Three packs available — Core / Elite / Supreme. See pack details on the checkout page. Indicative range: €1 497 to €4 997 (excl. VAT).
Learning Objectives	Acquire and reconstruct: design metric reconstruction pipelines from LiDAR, photogrammetry, NeRF and 3D Gaussian Splatting (Modules 1, 2). Engineer intelligent point clouds: build segmentation, classification and 3D deep learning systems on real datasets (Modules 3, 4). Architect agentic spatial AI: integrate LLMs, scene graphs and digital twin pipelines into production-ready systems (Modules 5, 6).
Target Audience	Engineers, data scientists, R&D developers, BIM managers and DeepTech entrepreneurs aiming for a senior Spatial AI role.
Prerequisites	Working Python and basic linear algebra. Watch the prerequisites primer →
Estimated Duration	Approximately 84 hours of focused work across the 6 modules (≈ 12 weeks at part-time pace). Fully asynchronous; modules are self-contained.
Access	Enrolment via the 3D Geodata Academy. A 14-day legal cooling-off period applies (see section 7). Minimum 14 days between enrolment and start.
Accessibility & Disability	All courses are open to learners with disabilities. A dedicated referent reviews each request to put the right pedagogical and technical adjustments in place. Referent: Dr. Florent Poux — howto@learngeodata.eu.
Contact	Dr. Florent Poux — howto@learngeodata.eu 3D Geodata Academy

A note from Dr. Florent Poux This is the program I wish I had when I started in 3D research. It is engineered to compress what usually takes a PhD plus three years of industry into a structured, project-driven path. Every module ends with a deliverable you can show to a hiring manager or a client. In the AI era, generalists vanish; spatial AI specialists win.

2. Technical Stack & Pedagogical Means

Software stack: Python 3.11+, NumPy, SciPy, Pandas, Open3D, PyVista, PyTorch, Laspy, Trimesh, Flask, Three.js.
3D tools: CloudCompare, MeshLab, RealityCapture, Meshroom, Blender.
AI stack: LangChain, foundation models (SAM, DepthAnything), Streamlit.
Hardware: any modern laptop (Win/Mac/Linux). GPU recommended for Module 4.
Infrastructure: proprietary LMS with 24/7 access, automated quizzes, progress tracking and digital course materials.
Modality: alternates theory and intensive hands-on practice — code, real datasets, software use and shippable project deliverables.
Theory resources: video lessons, written handouts, curated research articles and PhD-level references.

3. Program Structure

Module	Title & Focus	Project Deliverable
M1	3D Python Systems 3D maths, environment, multimodal viewers.	3D Data Processing Software.
M2	3D World Reconstruction LiDAR, photogrammetry, NeRF, 3DGS.	One-Click Reconstruction Engine.
M3	Smart Point Clouds Segmentation, RANSAC, SAM 3D, clustering.	HITL Labelling System.
M4	3D Deep Learning PointNet++, KPConv, 3D Transformers.	End-to-end 3D Classifier.
M5	Spatial AI & Agents LLMs, scene graphs, LangChain, reasoning.	Agentic Spatial Web App.
M6	Digital Twins & BIM Scan-to-BIM, IFC, topology, change detection.	Digital Twin Geometric Engine.

Module 1 — 3D Python Systems

Establishes a production environment and the 3D maths needed to build analytical micro-software.

Anaconda / Spyder environment and project structure.
3D data science as an operating system.
3D tools setup: libraries and automations.
Maths for 3D: vectors, dot products, coordinate systems.
3D representations: point clouds, voxels, meshes and conversions.
Multimodal 3D viewer development.
LiDAR-based building reconstruction primer.
Local Python analytical apps; LLMs as accelerators.

Project 1: multimodal 3D data processing software.

Module 2 — 3D World Reconstruction

Builds automated pipelines that turn physical environments into metric digital models.

3D reconstruction fundamentals.
Photogrammetric data acquisition on site.
Full reconstruction pipeline (SfM, MVS).
Python automation: image / video to 3D model.
Camera calibration and metric photogrammetry (GCPs).
LiDAR fundamentals and large-scale scanning.
Batch processes for 100% Python automation.
Monocular depth, image-to-voxel foundation models.
NeRF and 3D Gaussian Splatting; web/desktop sharing.

Project 2: one-click reconstruction engine in Python.

Module 3 — Smart Point Clouds

Applies geometric and unsupervised algorithms to extract meaning from raw spatial data.

3D latent space (voxels, mesh, CAD, 3DGS to point clouds).
Pre-processing and voxel-grid sampling.
Big-data strategies: chunks, tiles, streaming.
Smart point cloud architecture.
PCA and feature extraction; n-order RANSAC and ground detection.
Spatial structures: KD-Tree, Octree, search strategies.
K-NN and Euclidean clustering; region growing; DBSCAN/HDBSCAN.
HITL labelling systems; PostgreSQL/PostGIS spatial DBs.
Bounding-box extraction; semantic segmentation via 3D ML.
Multi-modal fusion; Segment Anything (SAM) for 3D.
Custom 3D labelling GUI.

Project 3: intelligent labelling system for massive point clouds.

Module 4 — 3D Deep Learning

Designs, trains and deploys neural networks for 3D classification and segmentation.

3D deep learning workflows and architectures.
ANN/CNN basics for spatiality; PyTorch custom classes.
PointNet: architecture, data preparation, training, inference.
ResNet and EfficientNet in 3D; CNN/R-CNN from scratch.
PointNet++ for indoor semantic segmentation.
KPConv on aerial LiDAR.
GrowSP (unsupervised + supervised).
Generative models for 3D; hybrid AI segmentation systems.
Local and cloud deployment of 3D DL apps.

Project 4: end-to-end 3D deep learning classifier.

Module 5 — Spatial AI / Agentic AI / LLMs

Integrates LLMs with 3D data to build autonomous agents capable of spatial reasoning.

Geometry, semantics and topology for AI.
Workflows from data science to spatial AI.
Unsupervised 3D object detection; HITL semantic injection.
Graph theory for 3D datasets; scene graph generation and characterisation.
Graphs as latent space for AI.
Agent architecture and LLM integration (LangChain).
From chat to agents: scoping, planning, reasoning.
File-system middleware, spatial indexing and transformation tools.
Prompt engineering for 3D spatial contexts.
Production: Streamlit chat apps and cloud deployment.

Project 5: agentic web app for spatial perception and action.

Module 6 — Digital Twins / Scan-to-BIM

Builds precise digital twins through advanced vectorisation and asset management.

Definition and stakes of the 3D digital twin.
Open standards: BIM, CIM, IFC, smart cities.
Open-source libraries; OpenData for context extraction.
Interactive platform and 3D data catalogue.
Static + real-time data layer integration.
2D/3D LiDAR vectorisation for urban modelling.
End-to-end Scan-to-BIM framework: geometric, semantic, topological engines.
AI-driven semantic modelling; standardisation, exports, interoperability.
Change-detection engine.

Project 6: complete digital twin creation pipeline.

4. Assessment, Certificate & Grading

The Spatial AI Architect program is a full program with project defence. Evaluation runs across the entire path:

Stage	Activity	Validation
Before the program	Positioning quiz (Python, linear algebra).	Informative — adapts mentoring intensity.
During the program	End-of-module quiz (×6).	Score ≥ 70 % per quiz.
End of the program	Final quiz (120 questions across 6 modules).	Score > 80 %.
End of the program	Portfolio submission — six project deliverables.	Reviewed against published rubric.
End of the program	Oral defence with Dr. Florent Poux (45 min).	Score ≥ 15 / 25.

Conditions to obtain the certificate

Validate the six end-of-module quizzes (≥ 70 %).
Validate the final quiz (> 80 %).
Submit the six project deliverables and pass the oral defence (≥ 15/25).

Grading scale

Certified — Pass: all criteria validated.
Certified with Merit: oral defence ≥ 18/25.
Certified with Distinction: oral defence ≥ 22/25 and quiz average ≥ 90 %.

Successful learners receive the Spatial AI Architect certificate and the Alumni digital badge.

Accessibility & disability: all evaluations can be adapted (extended time, alternative formats, oral or written substitution) on request to the disability referent howto@learngeodata.eu.

5. Program Results & Quality Indicators

The 3D Geodata Academy publishes its program indicators transparently. Figures below cover the Spatial AI Architect cohort and are updated at the end of each session.

Indicator	Current Result	Target
Number of enrolled learners	Data being consolidated (new program cohort)	> 10 / year
Satisfaction rate	Pending end of cohort	> 95 %
Success rate (certificate obtained)	Pending end of cohort	> 85 %
Drop-out / interruption rate	0 %	< 5 %
Professional placement rate	Pending end of cohort	> 80 %

Indicators updated at the end of each program session. Last update: April 2026.

6. Next Step

The Spatial AI Architect program is the most complete path we offer. It pairs the structured curriculum with direct mentorship from Dr. Florent Poux, the full course library (20+ courses), monthly analytics on the 3D spatial AI ecosystem, curated research papers and the private job board with reviews and notes on which roles are worth pursuing.

🛒 Secure your spot 🚀 Program details

7. Legal, Accessibility & Purchase Conditions

Digital accessibility

3D Geodata Academy is committed to making its content accessible to learners with disabilities. A dedicated referent oversees content accessibility (captioning, transcripts, screen-reader friendly layouts, alternative evaluation formats). Contact: howto@learngeodata.eu.

Access for learners with disabilities

Every program can be adapted upon request. A questionnaire at enrolment captures the pedagogical and technical adjustments needed. Disability referent: Dr. Florent Poux — howto@learngeodata.eu.

Payment terms — IP-protection caution & 14-day retraction

Because program content is delivered digitally and immediately accessible, the purchase is structured as follows to protect the intellectual property while preserving your legal right of retraction:

At checkout, the payment is collected as a caution / security deposit for IP infringement protection. LMS access is granted immediately.
The deposit is held for 14 calendar days, matching the legal cooling-off period for distance contracts.
If you exercise your right of retraction within those 14 days and have not consumed a substantial part of the content, the deposit is refunded in full.
At the expiry of the 14-day period, in the absence of a retraction request, the deposit is automatically converted into the actual payment.

By purchasing, the learner acknowledges that the program materials are original works protected by copyright. Any reproduction, redistribution or commercial reuse without prior written consent is prohibited.

Data protection (GDPR)

Personal data is processed for the sole purpose of delivering the program, tracking progress, issuing the certificate and providing customer support. Access, rectification or deletion: howto@learngeodata.eu.

General terms of sale

Full general terms of sale are available on request and joined to every order confirmation.