Senior Deep Learning (m/f/d)

Omegga GmbH

Job Overview

Location

Munich

Job Type

Full-time

Full Job Description

📋 Description

• Advance deep learning models through research iteration by developing and testing architectural improvements, training objectives, and optimization techniques, owning the full loop from hypothesis to experiment to conclusion and next iteration.
• Build rigorous evaluation frameworks and benchmarks by defining evaluation sets, establishing clear metrics (precision, recall, accuracy, calibration), and creating repeatable benchmark runs to ensure measurable and comparable performance improvements over time.
• Own end-to-end monitoring of model quality in production by setting up performance tracking, defining alerting signals for regressions, and building lightweight reporting systems to detect data shifts and model degradation early.
• Partner cross-functionally with data and engineering teams to improve datasets and labeling strategies, and collaborate with product and operations stakeholders to align on practical definitions of model performance and success.
• Architect problem-specific deep learning models by adapting modern architectures, losses, and training objectives to the unique structure of in-ovo sex detection tasks, treating published research as a starting point rather than relying on off-the-shelf solutions.
• Develop and maintain robust Python-based deep learning pipelines using PyTorch or TensorFlow, with strong discipline in experimentation, reproducibility, and model versioning.
• Apply deep mathematical understanding of modern machine learning—including probability, optimization, information theory, and statistical inference—to derive custom loss functions and training strategies rather than relying on default implementations.
• Utilize experiment tracking and model evaluation tooling (e.g., Weights & Biases) to drive measurement-based progress, ensuring all iterations are logged, analyzed, and actionable.
• Work in a high-ambiguity, fast-paced environment where real-world sensor and time-series data from industrial settings present frequent edge cases and dataset shifts.
• Translate research findings into production impact by collaborating with engineering teams to integrate models into the algorithm stack and ensure deployment readiness under real-world constraints.
• Contribute to improving model robustness and detection quality in noisy, real-world environments where false alarms and rare events carry high operational costs.
• Engage in continuous learning and innovation by exploring domain adaptation techniques, agentic AI workflows for experimentation, and efficiency improvements in transformer architectures.
• Maintain scientific rigor while accelerating experimentation through structured analysis, ablation studies, and test scaffolding, ensuring traceability and reproducibility of results.
• Operate within a mission-driven team focused on eliminating chick culling globally through non-invasive, AI-powered in-ovo sex detection technology.

🎯 Requirements

• MSc in Computer Science or related field with 4+ years of applied deep learning experience, or PhD with proven track record of taking research from idea to working system
• Strong understanding of optimizing modern neural architectures, particularly transformers, with practical experience in attention variants, efficiency improvements, and training stability
• Ability to architect problem-specific models by adapting architectures, losses, and training objectives to task structure, avoiding off-the-shelf solutions
• Strong Python deep learning stack experience (PyTorch, TensorFlow) including training pipelines, experimentation discipline, and reproducibility
• Deep mathematical knowledge of modern ML: probability, optimization, information theory, statistical inference; ability to derive loss functions
• Solid experience with experiment tracking and model evaluation tooling (e.g., Weights & Biases) and a measurement-driven approach to progress

🏖️ Benefits

• Fully furnished apartment within walking distance of the office for up to 6 months for relocation support
• Up to 50% remote work, depending on role and execution needs
• 28 vacation days per year, plus December 24th and 31st off
• Competitive, market-aligned salary linked to impact

Skills & Technologies

Python

TensorFlow

PyTorch

Senior

Remote

Degree Required

Ready to Apply?

Apply Externally

You will be redirected to an external site to apply.

AI Job Fit Analysis

Pro

See exactly how your profile matches this role — strengths, skill gaps, and what to do about them.

Omegga GmbH

Visit Website

About Omegga GmbH

Omegga GmbH is a German biotechnology company that has developed a non‑invasive, in‑ovo sexing technology for poultry eggs. Using Raman spectroscopy, the system can determine the sex of chicken embryos early in incubation, allowing hatcheries to identify and separate male eggs before they develop, thereby eliminating the need for post‑hatch culling of male chicks. The technology aims to improve animal welfare, increase hatchery efficiency, and meet growing regulatory and consumer demands for ethical poultry production. Omegga works with breeders, equipment manufacturers, and poultry producers to integrate its solution into existing incubation lines, and has secured funding from European venture funds and grants to scale its platform globally.

View Company Profile

Get more remote jobs like this

Subscribe to the weekly newsletter for similar remote roles and curated hiring updates.

Weekly remote jobs and featured talent.

No spam. Only curated remote roles and product updates. You can unsubscribe anytime.