- What core problem are we solving, and who is the target user? fast, safe, db to support every aspect of nlp, machine vision and machine hearing
- What are the absolute must-have features for a functional MVP? ability to store documents, efficiently store them in a format optimal for ai and ml ops. must support parallel reading and writing operations for multiple users/processes.
- What features can be postponed for later versions? start with .txt support only, but data formats need to be future-proof to easily integrate all other data formats.
- What are the main user stories and workflows? multiple ai/ml agents will be running and accessing the db simultaneously from one end of the ai/ml pipeline to the other. They may be running on seperate machines or the same machine.
- What platforms and devices must the MVP support? A variety of Ubuntu Linux machines.
- What integrations (APIs, third-party services) are essential at launch? None. This db should be self-contained with no external dependencies.
- What are the key performance and scalability requirements? Must be suitable for small home projects up to ultra-scale enterprise and scientific computing
- What are the security and privacy requirements for the MVP? The first version will have none, but it needs to be futureproof to be able to easily add in complete zero-trust best practice security features in the future.
- What data will the MVP collect, store, and process? The first version will support only text files. The second version will add other text document formats. The third version will also support machine vision for images. The fourth version will support machine hearing. The fifth version will also support video. The sixth version will add external db support. The seventh version will include security features.
- What is the simplest architecture that can support MVP needs? To be researched.
- What technology stack aligns best with project goals and team skills? Rust, Linux.
- What is the estimated time and budget to deliver the MVP? August 2025.
- What are the major technical risks, and how can we mitigate them? To be researched.
- What will success look like (metrics/KPIs) for the MVP? To be researched.
- What is the plan for testing and quality assurance? To be researched.
- How will user feedback be collected and integrated into future iterations? GitHub bug reports.
- What is the deployment and release strategy for the MVP? Open-source project on GitHub.
- Who are the key stakeholders, and how will communication be managed? AI and ML engineers and data scientists.
- What is the plan for post-MVP support and maintenance? Indefinite, ongoing development.
- How will documentation be handled for users and developers? On personal wordpress server.
RagPipeDB MVP Specification v0.0
Note: This document is a WIP and needs to be integrated into KB after completing version 1.