autarc is the OS for HVAC installers to install more heat pumps. 30% of Europe's CO2 emissions result from heating, and we are at the forefront of combating this environmental concern. Through LiDAR technology and AI, we reduce the overall project time of heat pump installations by 50%.

We're seeking an exceptional Machine Learning Engineer with expertise in computer vision to join our growing team. You will be our first ML Engineer, set up our entire ML infrastructure and play a critical role in developing innovative solutions to streamline heat pump installations.

Responsibilities:

• Develop and implement machine learning models. The core will be a computer vision model based on camera and LiDAR data.
• Design, build, and maintain the ML infrastructure using Roboflow
• Preprocess and analyze large datasets
• Train and fine-tune machine learning models
• Collaborate with cross-functional teams to integrate ML solutions into the platform
• Monitor and improve the performance of deployed models

Qualifications:

• Strong proficiency in machine learning and computer vision techniques
• Experience with ML frameworks such as TensorFlow, PyTorch, or similar
• Optional: Experience with Roboflow
• Optional: Experience with Apple’s ARKit or any other augmented reality tools
• Proficiency in Python and related ML libraries
• Familiarity with LiDAR technology and data processing
• Experience in setting up and managing ML infrastructure
• Preferably based in Berlin, with a willingness to work from our office close to Rosenthaler Platz

What we’re offering:

• Office in Berlin-Mitte: Our office is located near Weinbergspark in Berlin-Mitte.
• Equipment: You have the freedom to choose your own equipment (within a budget range).
• Annual Team Retreat: Let’s enjoy an annual retreat with the entire team.
• Mobility Package: Choose between a JobRad and Dance Bike for your mobility.
• Fitness Package: Choose between Urban Sports and Classpass for your fitness.
• Collegial and Familial Team: A collegial and familial team welcomes you, characterized by an open company culture.
• Part of the Energy Transition: Be a part of the energy transition and directly experience the impact of your daily work.

Our app has two interesting technical challenges: Multiple custom machine learning models, one of them analysing camera and LiDAR data in real time and an offline-first React web app.

The core part of our product is a room scanning feature, which allows users to capture a geometrical representation of their customer houses. Currently, we are using an existing solution for the actual scanning part, but we are quite limited with that, which is why we need to build our own. For this we are training a machine learning model that combines camera and LiDAR data to detect objects and dimensions in real time. We then use augmented reality to visualise the scan results (also in real time). To make this work we need you!

Next to this, our app has to work entirely offline, including the machine learning model. For this we run a local database inside of our app and sync that to a Supabase database in the cloud when online.