Purpose & people
About FridayLocalAI
Useful intelligence. Human direction.
FridayLocalAI is a local-first AI platform for people and organizations that want practical workflows on infrastructure they control.
More than a model
The application brings together conversations, projects, knowledge workflows, model controls, and governed capabilities. The runtime, stored evidence, policy, and operator define system truth.
Built around the work
Friday’s Workshop gives those capabilities a shared interface. Its development emphasizes traceability, explicit boundaries, and maintainable operation.
Explore this topic →Part of Hubbard Genesis
FridayLocalAI is a project of Phronesis Intelligence, a DBA of Hubbard Genesis Corporation. Explore the platform, read the documentation, or contact us about your intended use.
Explore this topic →Learn by doing
FridayLocalAI is intended to support ongoing work: organize a project, build context, ask focused questions, and check the results. The guides on this site use that same approach. Reading material can inform how you think about AI, but no purchase is required to follow the documentation.
Ask Friday to guide you
In an authorized FridayLocalAI conversation, try:
Help me decide whether FridayLocalAI fits my intended workflow. Ask about my goals, information sensitivity, operating environment, and need for connected services. Separate established capabilities from items I should confirm with the project.
Tell Friday what you can see and which step you have reached. If the instructions do not match your screen, say so; ask for the relevant Help topic or contact your operator. Friday cannot be assumed to see your screen or know your current settings.
The problem behind the project
AI work often begins with a single question but soon grows into a collection of sources, drafts, decisions, and follow-up tasks. Without a clear structure, users can lose track of which conversation holds the latest decision or whether an answer actually used the intended material. FridayLocalAI approaches that problem as a workspace problem as well as a model problem. The project brings together conversations, organization, context, model controls, and governed capabilities. Its purpose is to support work that people can return to, inspect, and direct. A useful response matters, but so does knowing where the work belongs, what evidence supports it, and which action actually occurred. Those concerns shape both the application and the guidance on this website.
Why local-first is a design choice
Local-first AI gives the operator a direct role in decisions about infrastructure, models, storage, and access. It does not mean every possible feature is disconnected or that every deployment has identical privacy characteristics. Optional connections and provider-based capabilities need to be understood in the context of the actual workflow. FridayLocalAI therefore describes local operation alongside its responsibilities. Someone must maintain the environment, review changes, and provide a support path. Users need to know which choices are theirs and which belong to the operator. The website does not replace that discussion with a universal hardware claim or a promise that all information remains within a particular boundary regardless of configuration.
Friday’s Workshop gives the work a home
Organizer helps users navigate projects, folders, conversations, and messages. The conversation provides the working exchange. Dashboard brings context and related controls into view, while Media Studio provides a dedicated surface for enabled creative workflows. Agents, models, Skills, Tools, and Memory Notes have different roles within that environment. Those distinctions are intended to be understandable. A user can ask Friday to explain a control, work through a small task, or help prepare a support report. Guidance is most useful when it remains connected to visible actions and results. If a save, source review, or generation is claimed, the relevant record or artifact should be available for inspection rather than existing only as a sentence in a reply.
Human direction remains central
The project treats the model as one component of the system. Runtime behavior, stored evidence, permissions, and operator authority determine what is actually available and allowed. A prompt can express a goal, but it does not independently create filesystem access or make an external action appropriate. A specialized Agent does not remove the need to review important output. For users, this translates into practical habits: state the task clearly, choose the right context, ask about uncertainties, and inspect the result. For source work, compare claims with the original material. For durable guidance, verify saved notes and scope. For Tool actions, check the resulting state. These habits support useful AI assistance without confusing confidence or fluency with proof.
The role of learning and documentation
The learning library offers background on AI, engineering, measurement, and decision-making. Its books and resources are not a required purchase path or a claim of endorsement by the authors. Linked products use the website’s Amazon Associates module with disclosure, and their retailer listings should be checked for the intended edition and format. The application guides focus on doing a small task well. They explain first-run checks, conversations, attachments, model choices, memory, and troubleshooting. Consumer PDFs collect guidance for convenient reference. The objective is to help users ask better questions and recognize a successful result, not to replace deployment-specific instructions or present every documented concept as universally enabled.
Who operates the project and how to begin
FridayLocalAI is a project of Phronesis Intelligence, a DBA of Hubbard Genesis Corporation. The website’s HGC links provide navigation to the related corporate sites. This affiliation identifies the project; it does not imply that every linked site shares the same product, account system, or operating terms. If you want to explore the application, contact the project with your intended workflow and environment. Early access is arranged by inquiry, with no public installer or current store purchases. If you already have access, follow the operator’s instructions and begin with a non-sensitive practice task. A productive introduction connects your goal with the actual capabilities, context, and review process available in the deployment you will use.
About this documentation
These pages were prepared with AI assistance using inspected application source and project direction. Examples are instructional, and the product descriptions distinguish source-supported behavior from deployment-specific availability. If your installation differs, the operator’s verified guidance should resolve the discrepancy before you continue.
The publications behind the project
David Alan Hubbard’s Phronesis Intelligence series explores the experience, methods, and architectural questions behind the project. What We Learned Together examines AI collaboration and human judgment. After the Prompt develops practical habits for questioning and improving AI-assisted work. Practical Wisdom in the Machine explores ownership, authority, continuity, and governed systems. The publications are a source of context and working ideas; they do not replace current release documentation or verification of a particular installation. The reading guide connects those ideas with concrete tasks you can try during early access.