Embedded Files
About the company
Spatial Intelligence is building the future of 3D automation. Our mission is to be the first to build generalisable “Spatial Intelligence” for 3D automation - giving machines the ability to understand, construct, and interact with the 3D environments. With Spatial Intelligence, we will unlock new possibilities for automation across gaming, robotics, architecture, and beyond.
We aren’t building another 3D tool - we’re building the core intelligence layer that will power them all…
We started by asking ourselves: “what if we built the intelligence behind tools that could make 3D scenes in one click?”
About the role
We are seeking an ambitious and impact-driven Research Engineer to join our innovative team in building language-driven systems for 3D scene generation. You'll bridge the gap between our research and engineering teams - driving the implementation and development cutting-edge research into industry-leading products. Day-to-day, you'll be combining large language models, computer vision, and 3D AI to create intelligent systems that can understand, generate, evaluate, and manipulate complex environments from natural language descriptions.
Example focus areas include:
Text, graph and neural scene representations
Semantic, spatial and relational scene modelling
Text- and image-guided 3D scene generation
Multi-modal embedding, retrieval and RAG
LLM-driven 3D scene parsing and evaluation
Responsibilities
Design and implement novel algorithms for language-driven 3D scene generation and manipulation.
Develop scene graph generation systems that create structured representations of indoor environments.
Research and implement text-to-image and text-to-scene generation pipelines.
Create LLM-as-a-judge frameworks for evaluating generated scenes and ensuring quality control.
Experiment with prompt engineering techniques to optimise model performance across different scene generation tasks.
Integrate foundation models and cutting-edge research approaches into production-ready systems.
Ideal candidate
Leader/owner mindset - a future leader who leads with vision, responsibility, and togetherness.
Ambitious, curious and impact driven - obsessed with acquiring knowledge, deepening understanding, and translating findings into real-world outcomes.
Early-stage fit - compelled by hands-on, high-trust roles within tight-knit, lean teams. Focused on quickly iterating, deploying, and scaling meaningful software based on feedback.
A grower - strong growth curves in past roles, progressing quickly through their careers.
Ideal qualifications
STEM MSc/PhD - Computer Science, Machine Learning, Computer Vision, or related field.
NLP/GenAI experience - worked with large language models, RAG/semantic search, prompt engineering, and transformer architectures.
Computer vision proficiency - image embedding and multi-modal models, exposure to text-to-image models.
Python proficiency - experience writing clean, testable Python code with exposure to modern Python development and software engineering practices (e.g. APIs, GitHub, Code reviews).
(Optional) Cloud Platforms - hands-on experience with Azure and/or AWS.
(Optional) Containerisation - proficiency with Docker and container orchestration.
(Optional) Publications in top-tier AI conferences or demonstrable product delivery experience in one or more areas: multi-modal AI, 3D computer vision, generative models, or scene understanding.
What we offer
Package - £45k-£75k base salary (depending on experience), 0.5-1.0% equity. (Equity-weighted to attract owners. We're an ambitious company positioned for explosive growth, and want others who will call our company and our mission "their own".)
Frontier AI projects - a chance to define a new field at the intersection of computer vision, natural language processing and 3D automation.
Technical leadership opportunity - architect systems that will define the future of AI-driven 3D automation.
The chance to publish and speak - if you want visibility, we’ll support it.
An industry-leading product team to support your impact - in a year’s time, we expect your research and engineering to be at the centre of category-defining games (e.g. narrative-driven infinite game worlds that build themselves around the player).
How to apply
We're looking for someone exceptional who wishes to achieve exceptional things with others. Because of this, we’re very open-minded about backgrounds - we didn’t all come from conventional paths and don’t all fit the template! If you believe you’re a fit, even if you don’t meet 100% of the criteria, please reach out.
Send a short intro + your GitHub, portfolio, or CV to hello@spatial-intelligence.co.uk
We believe it’s time AI understood the 3D world. If you do too, come and tell us why this mission speaks to you!
Page updated
Google Sites
Report abuse