Dydema
Dydema is building a non-invasive brain-computer interface (BCI) headset and the AI models that interpret brain signals, aiming to make thought-based computer control as routine as using a keyboard. The company offers a developer API that tests and hosts EEG classifiers on user-provided data, with a focus on real-world, non-medical applications. Their current technology can distinguish a limited set of thoughts and is being designed for everyday use, with a goal of making BCI communication faster than speech.
- Artificial Intelligence
- Developer Tools
- Software Only
- Hardware
Funding
Raised to date
$500KRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Founder details are not available yet.
Product
Problem
Developing neurotechnology applications requires both sophisticated AI models that can interpret brain signals and large volumes of high-quality neural data to train those models. Most organizations lack the in-house recording infrastructure and expertise needed to collect and process EEG and other neurophysiological data, creating a significant barrier to entry in the neurotech space.
Solution
Dydema provides a dual offering: custom AI models that make sense of brain signals, and the recorded neural data required to train them. The company delivers custom EEG and other neurophysiological datasets on demand, allowing organizations to build neurotech applications without investing in their own recording equipment or data collection pipelines. By combining model development with data provisioning, Dydema offers an end-to-end solution that reduces the technical and logistical hurdles of entering the neurotech field. The company is also developing its own headset with cheaper sensors and a model API, aiming to make BCIs a non-medical, everyday tool.
Target Audience
Primary customers are organizations and developers building neurotech applications who need AI models for brain-signal interpretation and the training data to support them, including companies without existing neural data recording capabilities. The developer API targets researchers and developers who have recorded EEG trials and need a hosted classifier.
Features
- Custom AI models designed to interpret and analyze brain signals
- On-demand EEG and neurophysiological dataset creation tailored to client specifications
- Full data recording service that removes the need for in-house recording infrastructure
- Combined model and data offering that streamlines the neurotech development workflow
- Developer API (Wizard) that tests licensed pretrained models on user-labeled trials and hosts the best-performing one as a classifier
- API supports any 10-10 or 10-20 channel layout, sampling rate, and trial length, with tasks including ERP, motor imagery, and sleep staging
- Hosted task models provide per-trial labels and class probabilities, with a validation report and measured accuracy
- Research-use-only API with a one-time creation fee and monthly hosting fee