Senior Applied AI Researcher— Sovereign AI Research

Description

Every nation has data. Few can protect it. Fewer still can act on it.

Dream is the sovereign AI and national cyber-defense company for governments.

We help nations secure their most critical systems, connect fragmented information at a national scale, and turn their most sensitive data into decisions, all fully sovereign.

This is more than a job. It's a Dream job, where you'll work at a global scale alongside some of the best AI researchers, cyber operators, and government experts in the world.

We're building AI that nations own and control, deployed where almost no one else can operate. ingesting and structuring complex data, and driving practical actions that can literally impact the lives of billions of people around the world. This role helps make that real.

The Dream Job

Nations are waking up to a hard truth: critical intelligence infrastructure built on hyperscaler black boxes isn't a solution it's a dependency. The DREAM Sovereign AI Research Group exists to answer that differently.

We're not fine-tuning what already exists. We're rethinking the models architecture from the ground up modular, composable, and built with compute governance as a first-class design constraint, not an afterthought.

We operate under real-world constraints. The interesting problems live at the intersections of disciplines. That's where we operate.

This is a hands-on research role embedded within a team of senior researchers. Day to day includes training models, running benchmarks, synthesize data. The researchers you'll work alongside will challenge you technically and invest in your growth.

The Dream-Maker Responsibilities

Open Research Tracks

Familiarity with at least one is expected:

• Computer Vision: object detection, segmentation, multimodal grounding, vision-language models, contrastive and self-supervised representation learning, low-resource and few-shot visual recognition.
• NLP / Speech: LLMs, NERs, relation extraction, span-based and generative IE, semantic textual similarity, multilingual and cross-lingual transfer.
• Reinforcement Learning: MDPs, POMDPs, model-based and model-free RL, Online Offline methods, reward modeling, sim-to-real transfer, compute-aware planning.
• Graph Learning: GNNs, graph clustering, community structure, generative methods, knowledge graph embeddings, dense and sparse semantic retrieval.
• Optimization: convex and nonconvex optimization, constrained and Lagrangian methods, combinatorial and integer programming, knowledge distillation (response, feature, and relation-based), test-time optimization, Bayesian optimization, resource-aware inference.
• Representation Learning: contrastive learning, self-supervised and unsupervised pre-training, disentangled representations, metric learning and embedding spaces, cross-modal and multimodal alignment, meta learning (hypernetworks), transfer learning and domain adaptation, probing and interpretability of learned representations, world models.
• Neurosymbolic AI: neuro-symbolic integration, differentiable theorem proving, inductive logic programming (ILP), probabilistic soft logic (PSL), causal inference and structural causal models (SCMs), programmatic and compositional reasoning

Responsibilities:

• Train and evaluate models across research tracks, iterating fast while documenting rigorously.
• Build and maintain benchmarking pipelines and evaluation suites.
• Curate, structure, and preprocess datasets; contribute to synthetic data generation workflows.
• Run ablations and controlled experiments to support research hypotheses.
• Reproduce and stress-test results from recent literature relevant to the group's work.
• Collaborate across tracks and with engineering teams through to production handoff.

The Dream Skill Set

• MSc in Computer Science, Electrical Engineering, Mathematics.
• Strong academic record with hands-on experience / thesis research.
• Proficiency in Python and at least one deep learning framework (PyTorch preferred).
• Comfort with the full data lifecycle: sourcing, structuring, cleaning, and transforming raw data into training-ready assets

We work on hard problems where data is scarce, labels are noisy. We expect researchers who don't wait for better conditions, who will reformulate the problem when needed, devise novel strategies to generate or simulate data, and find signal where others see noise. If your instinct when blocked is to push harder rather than halt, you'll thrive here.

Never Stop Dreaming...

If you think this role doesn't fully match your skills but are eager to grow and break glass ceilings, we’d love to hear from you!