Perception.Intelligence.Action.

Technical research on how intelligent systems perceive, reason, and act.

Neuroide is a startup research lab focused on the full stack of autonomy. The work spans physiological signals, machine learning, generative AI, and robotics, with an emphasis on systems that hold up outside controlled demos.

Read Featured Notes About the Lab
Core Areas

Core Areas

Neuroide is organized around the three layers that make autonomous systems work: perception, intelligence, and action.

The lab treats autonomy as a systems problem rather than a model category. Reliable behavior starts with perception, improves through learning and reasoning, and only matters when it survives contact with real environments, constraints, and failure modes.

Perception

Signals, sensing, calibration, and turning noisy measurements into usable information.

Intelligence

Representation learning, reasoning, uncertainty, and decision-supporting models.

Action

Control, robotics, deployment, and systems that execute under real constraints.

Perception / Signals

Signals

Research on physiological sensing, time-series structure, filtering, multimodal fusion, and the realities of noisy measurement.

Biosignals EEG EDA ECG PPG DSP
Intelligence / ML

ML

Work on inference, optimization, representation learning, and robust models that support real decisions.

Inference Optimization Representations Robustness Evaluation
Intelligence / GenAI

GenAI

Structured notes on reasoning systems, generative models, multimodal interfaces, and practical model behavior.

Generative Models Reasoning Multimodality Language Systems Serving
Action / Systems / Robotics

Robotics

Systems thinking on control, planning, middleware, simulation, and embodied execution outside the lab.

Control Planning Simulation Middleware Validation