Research Scientist Trustworthy AI and Model Security
Ulm, Germany | Hybrid / On-Site
About the Role:
As a Research Scientist in Trustworthy AI and Model Security, you will be responsible for ensuring the safety, reliability, and ethical integrity of ACC Group’s Financial AGI systems. Your work will focus on model robustness, adversarial evaluation, interpretability, and security-driven AI research to ensure our models can operate transparently and safely in high-stakes financial contexts.
You will develop evaluation frameworks, adversarial testing suites, and interpretability tools that help our models reason responsibly under uncertainty. This position sits at the intersection of AI safety, governance, and applied security engineering.
Key Responsibilities:
- Research and develop methods for detecting and mitigating bias, drift, and adversarial vulnerabilities in Financial AGI models
- Design and conduct stress tests and red-teaming evaluations to assess robustness and resilience under adversarial conditions
- Build frameworks for model explainability, auditability, and accountability in regulated environments
- Collaborate with compliance, security, and risk management teams to align model design with FINMA, BaFin, and AI governance standards
- Lead initiatives in secure model deployment, differential privacy, and cryptographic assurance for sensitive data
- Develop tools to evaluate alignment between AI outputs and human financial judgment
- Establish formal evaluation benchmarks and metrics for trustworthiness, safety, and ethical compliance
- Publish and present safety and governance findings internally and externally to enhance transparency and trust
Requirements:
- PhD or equivalent research background in Machine Learning, Computer Security, or AI Safety
- 5+ years of experience in AI robustness, adversarial machine learning, or model evaluation
- Strong programming and research skills in Python, PyTorch/TensorFlow, and data-driven security frameworks
- Expertise in adversarial training, model verification, interpretability, or risk-sensitive reinforcement learning
- Understanding of privacy-preserving techniques (differential privacy, homomorphic encryption, federated learning)
- Experience working in regulated or high-assurance environments (finance, defense, or critical infrastructure)
- Fluent in English; German is an advantage
Preferred Qualifications
- Passionate about AI safety and the ethical use of intelligent systems
- Rigorous scientific thinker with strong attention to reliability and reproducibility
- Comfortable navigating complex socio-technical challenges that blend AI research, governance, and compliance
- Driven by the mission to make AI systems transparent, secure, and beneficial for society
Why Join ACC Group
At ACC Group, we pursue ambitious goals — not incremental improvements. We are building the foundation for the next era of finance, and that requires people who thrive on innovation, precision, and curiosity.
Here, you’ll work alongside world-class engineers, researchers, and financial experts who share a common purpose: to create technology that redefines how the world understands and manages wealth. Our culture values initiative and independent thinking — every idea matters, every contribution shapes the outcome.
We offer an environment that combines the agility of a startup with the standards of a regulated financial institution. You’ll find the freedom to experiment, the responsibility to own your work, and the support of a team that values excellence over hierarchy.
Our people are our greatest investment. That’s why we provide competitive compensation, long-term incentives, and the opportunity to grow with a company that’s leading one of the most profound transformations in modern finance.
Join us — and help build the intelligence layer that will power the future of global markets.
About ACC Group
ACC Group AG is an AI-native fintech and asset manager shaping the future of asset and wealth management. We develop proprietary AI systems that transform data into actionable insights, enabling adaptive, data-driven, and transparent investment decisions. Our goal is to build financial ecosystems that continuously learn, proactively manage risk, and create sustainable long-term value.
Our research focuses on advancing Financial AGI (artificial general intelligence) designed for the complexity of global markets. These systems understand context, reason about uncertainty, and make decisions across multiple asset classes with the depth and precision of human expertise.
By combining deep learning, quantitative research, and modern infrastructure, we aim to make investing more precise, efficient, and accessible.
Our people are at the heart of everything we do. We believe that progress stems from curiosity, collaboration, and the courage to explore what's next. Together, we cultivate a culture that values excellence, rewards initiative, and brings together diverse perspectives toward one shared vision: advancing the way the world manages wealth.
Compensation
Competitive compensation package including base salary and performance-based bonus.We offer a comprehensive set of employee benefits designed to support your wellbeing and professional growth — including health coverage, continuous learning programs, flexible working arrangements, and access to modern tools and technology.