Alljoined
San Francisco, US Β· 5 known investors
Alljoined develops neural decoding technology using machine learning to enable computers to interpret human brain activity, aiming to enhance human capability, mental health, and interpersonal understanding. The company applies computational neuroscience to brain-computer interfaces for research and practical applications.
Also known as Alljoined
Founders & leadership
Investors Β· 5
Company profile
researched Aug 2026Alljoined is a neural decoding company developing machine learning models that extract semantic content and thought patterns from brain signals. Its research targets reconstruction of seen images, emotion decoding and, longer term, more complex cognitive processes including abstract reasoning, inner speech and intentional planning. The company frames its objective as mapping the space of human consciousness and improving how people understand themselves and interact with technology.
The work combines large-scale neural datasets with model development. Published outputs described by the company include ENIGMA, a multi-subject model that reconstructs seen images from EEG signals and adapts to a new user in minutes rather than hours by learning a shared representation across brains; a public EEG-image decoding dataset; MindEye2, a multi-subject fMRI-to-image architecture reported to train 40 times faster than prior state of the art at comparable performance; and a dataset for complex mental imagery reconstruction from fMRI. The company presents side-by-side comparisons of its EEG-based image reconstructions against reconstructions from other published methods.
Traction
Publicly stated outputs include peer-reviewed papers accepted at ICML and CVPR, the ENIGMA multi-subject EEG decoding model, and the release of a public EEG-image decoding dataset covering 20 subjects with over six hours of recordings each.
Latest developments
The company describes ENIGMA, a new multi-subject EEG-to-image decoding model, and is recruiting across computational neuroscience, machine learning research, software engineering, data infrastructure and executive support functions in San Francisco.
βΈFull profile β market position, technology, go-to-market, geography
Market position
Alljoined positions itself as a research-driven participant in brain-computer interface decoding, citing peer-reviewed publications, including papers accepted at ICML and at CVPR (as a highlight paper), and benchmarking its EEG image reconstructions against other published academic methods.
Technology
Alljoined's models operate on evoked neural responses, using both EEG and fMRI data. The ENIGMA model performs multi-subject EEG-to-image reconstruction by learning a shared "visual language" across brains, which the company says reduces per-user calibration from hours to minutes. MindEye2 is a multi-subject fMRI-to-image architecture reported to train 40x faster than the previous state of the art while matching its performance. The company also builds large-scale neural datasets, including one it describes as the largest public dataset for EEG-image decoding, with more than six hours of recordings per participant across 20 subjects collected on consumer-grade hardware, and models neural scaling laws on it, reporting log-linear scaling. A separate dataset covers complex mental imagery reconstruction from fMRI.
Go-to-market
The company's stated orientation is toward research and toward making practical brain-computer interfaces feasible for individual users, rather than a named commercial customer segment.
Geography
Operations and all advertised roles are based in San Francisco.
Compiled by commissioned research from 1 cited public sources β announcements, filings, and press listed under research sources below.
Key figures
latest reportedCompany-reported or press-reported figures, each dated to when it was claimed β not independently audited.
Timeline Β· 4
launches, deals, and filingsAlljoined introduced ENIGMA, a multi-subject decoding model that reconstructs seen images from EEG signals, learning a shared visual representation across brains so it can adapt to a new user in minutes rather than hours.
Alljoined released what it describes as the largest public dataset for EEG-image decoding, with 6+ hours of recordings per participant across 20 subjects collected using consumer-grade hardware, and modeled neural scaling laws showing log-linear scaling.
A dataset for complex mental imagery reconstruction from fMRI was accepted to CVPR as a highlight paper.
MindEye2, a multi-subject fMRI-to-image architecture, was accepted to ICML.
Dated company events from announcements, filings, and press; legal rows summarize public dockets and regulator releases.
βΈResearch sources Β· 1
primary sources listed
- Alljoinedalljoined.com Β· web
1 public sources were cited for this profile; the first-party ones are listed here.
Frequently asked questions
- What does Alljoined do?
- Alljoined builds AI models that decode images, emotion and thought patterns from brain signals such as EEG and fMRI.
- Who are Alljoined's investors?
- Alljoined's investors include Angel Collective Opportunity Fund, Compound, Evolution VC Partners, First Spark Ventures, Llc, Transpose Platform Management.
- Where is Alljoined headquartered?
- Alljoined is headquartered in San Francisco, US.