3D Reconstruction Engineer

Deepomatic's mission is to assist in the deployment of critical infrastructures necessary for digital and energy transitions through image recognition technology.

We develop software that, using AI and image recognition, automatically analyzes photos taken in the field by technicians and provides them with real-time feedback on the quality of their work, detecting possible defects and allowing them to fix issues before leaving. This enables projects to be completed right the first time, avoiding the need for a return visit, saving unnecessary travel and carbon consumption, accelerating infrastructure deployment, and improving the customer experience.

After securing a strong base of prestigious clients in the telecommunications industry, we are now targeting a much larger market with the goal of redefining how critical infrastructures are deployed and maintained. Joining Deepomatic means working at the intersection of crucial topics in our daily lives (digital and energy transition) and state-of-the-art technology, with a very tangible impact already at scale (20 million field interventions analyzed each year, and 30,000 technicians using our platform daily).

The Role

Our product offering includes a mobile experience that integrates custom-trained deep learning models with the tools used by technicians in the field. We would like to develop a new product line that would allow the precise localization of assets in space from video footages. The goal would be to combine structure from motion technologies with sensor fusion to obtain the best possible localization accuracy with the loosest hardware requirements. You will be the first member of a future 3D reconstruction team, in close collaboration with the CTO.

Within 1 month you will:

• Learn everything there is to know about the business, what is the strategy and the vision that drives the product development

• Meet the whole team to get to know them

• Define the R&D direction that has the best chance to lead to successful POC

Within 3 months, you will:

• Implement a first version of the POC.

• Define directions to improve reconstruction accuracy with other data (inertial, LIDAR, monocular depth estimation from deep learning)

Within 6 months your will :

• Collaborate with the product, marketing and commercial teams to freeze the hardware requirements of the targeted solution based on your POC

• Move on to the MVP implementation with the objective of developing a minimal yet production ready solution.

• Start investigating how the technology can serve other adjacent use cases.

• You have expertise in 3D reconstruction

• You have knowledge in GIS data formats

• You have knowledge in back-end development

• Having knowledge in sensor fusion is a plus

• You are known for your strong and positive interpersonal skills.

• You are proficient in English at a professional level.

If you don't meet 100% of the criteria above, you can still apply and explain why you believe you're the right person for the role!