Machine Learning Researcher (w/m/d) - Hybrid

Machine Learning Researcher (w/m/d) - Hybrid

Join to apply for the Machine Learning Researcher (w/m/d) - Hybrid role at dida

Responsibilities

• Work on machine learning research projects, addressing methods to improve performance in relevant applications such as computer vision or natural language processing, and general methodological questions (e.g., meta learning, hyperparameter search, efficient training of neural networks).
• Transfer theoretical insights to practical use cases within research projects with the goal of setting new industry standards.
• Publish results in relevant journals or conferences such as ICML, NeurIPS or ICLR.
• Foster strong connections to academia, keep up to date with advances in machine learning research and actively take part in the company’s collaborative learning and development environment.
• Support the dida team in applying cutting‑edge machine learning algorithms.

Qualifications

• MSc or PhD in mathematics, physics, computer science or a related field.
• Published academic papers; machine learning related papers are a plus.
• A creative mind that likes to solve problems.
• Grounded knowledge about machine learning and deep learning.

Why us?

Our team has a solid background in mathematics and statistics. We offer flexible working hours (full and part time) and a nice office with good coffee in Berlin Schöneberg. We prefer a hybrid work model but are open to remote work. We believe in science and support publishing your research results.

Projects we have worked on:

• Estimate the amount of solar panels that fit on a roof (computer vision): Given a satellite picture and a ground image of a house, automatically detect elements of a roof to determine how many solar panels fit on it, involving 3D inference from 2D pictures.
• Detect, classify and assess legal effectiveness of text paragraphs (NLP): Automatically analyze thousands of legal documents to classify paragraphs and check their legal effectiveness, involving scanning to text and labeling schemes.

dida is a machine learning software company tackling applied problems for different customers by using latest scientific advancements, especially in deep learning. Our research‑oriented thinking can help solve real‑world problems more efficiently.

At dida, we stand for equal opportunities, regardless of gender, nationality, ethnic background or disability. We encourage everyone, especially women, people of color and people with disabilities, to apply.