Machine Learning Researcher
Dubai, United Arab Emirates · Full Time
Be the first to apply
- Experience
- Any
- Salary
- —
- Openings
- 1
- Posted
- 3 days ago
- Work mode
- In office
- Education
- PhD or equivalent
- Resume
- Required to apply
Where you'll work
Sign in to tell us what does and doesn't work for you here — it sharpens every match we show you.
Job description
About Flashmind Labs
Flashmind Labs is dedicated to creating foundational AI models that directly interact with real-world data and evolve beyond the limitations of their initial training sets. These models prioritize accuracy, remain unconstrained in their truth-seeking process, and are tailored to make impactful decisions.
Our team consists of specialized researchers and engineers based in Dubai, committed to innovating the future of learning systems.
Role Overview
We seek a passionate researcher to expand the potential of foundation models optimized for veracity. This involves pioneering new training objectives, developing architectures that intrinsically manage calibration and uncertainty, and establishing evaluation systems focused on empirical results.
Key Responsibilities
- Develop innovative training goals focused on ensuring factual correctness and model calibration.
- Create model architectures capable of inherently managing uncertainty and distributional variability.
- Design and implement training processes aligned with empirically validated outcomes.
- Build evaluation frameworks to assess calibration accuracy, performance on rare events, and robustness to distributional changes.
Minimum Requirements
- PhD or comparable experience in machine learning, computer science, or related disciplines.
- Strong proficiency in Python programming and PyTorch framework.
- Capability to independently design, conduct, and interpret experiments.
- Familiarity with large-scale model training and use of accelerator hardware environments.
Preferred Qualifications
- In-depth knowledge in one or more areas such as alignment techniques (e.g., RLHF, DPO), model calibration, loss function engineering, or architectural design.
- A proven track record of contributions to advanced research, including publications or leading model developments.
- Experience with distributed large-scale training systems.
- Academic or practical background in epistemology, calibration theory, or decision sciences.
Minimum education
Doctorate