Skip to content
View Sayca-Jferone's full-sized avatar
🚀
Shell to Singularity || LEARN · AUTOMATE · ITERATE
🚀
Shell to Singularity || LEARN · AUTOMATE · ITERATE

Highlights

  • Pro

Block or report Sayca-Jferone

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Sayca-Jferone/README.md

🫡 Welcome, fellow human

I build Systems the way I think them: from the fuzziest intent to a living, auditable object,
with AI agents as peers under my review, or alone.

⚖️ Ethics | English v0.1 | French v0.1

🛠️ Methodologies

DRAFT | My methodology, opened

My own working methodology, refined and made public: the sequence by which I approach reality, specify a System and compose it. Five dimensions (ideas, rules, design, build, field feedback) in one .draft/ folder that humans and AI agents can audit and inherit. Use it to see how I work, or to work that way yourself.

HEXIS | Memory for AI agents

A portable memory and routing layer for my Claude agents, across every machine I work on: what has been reasoned once is stored as skills, facts and memories, so no reasoning is paid for twice. Private R&D, for now.


🧠 Posture

0️⃣ - No standard limitation

I do not reject a system or a method because it is exogenous or unconventional: standard judgements are what keeps rupture solutions from being found.

1️⃣ - System modeling by contracts

Every unit gets a written contract before it is built: what it owns, takes, does, guarantees, and how it fails. Flows and invariants, so logical blocks assemble into a System that scales.

2️⃣ - Crystallization of cognition

Research on automating and scaling cognition: tools that augment what a human can do alongside AI peers, and that keep the quality extracted from every block of tokens spent.

3️⃣ - Multi-channel communication

Humans and AI agents talk through self-documenting architectures: DRAFT Matrices, FMBOA requirements, BIOPGE contracts, Markdown, vector databases.


</> Tech stack

Most explored
Low level C11, C++23, Rust, Unix, Shell
Infrastructure Docker, Linux, Git, Cloudflare
Autonomous systems Python, LLM, APIs, VectorDB
Web & deployment Astro, PHP, Cloudflare services

⚙️ Activities

Latest Description
42 Next curriculum 🇫🇷:RNCP-7 🇪🇺:EQF-7 Intensive computer science studies. Proving ground for DRAFT and for team work.
DRAFT My methodology as an open-source method: a five-dimension contract for any System, any domain, any agent, any skill level.
NOESIS WIP Portable epistemic substrate engine: indexes, crystallizes and retrieves structured knowledge across inference cores. Better knowledge bases, lower inference cost.
PACT (archive) DRAFT's precursor: an attempt to reunify formal methods (B Method, VDM) whose completeness discourages individual adoption. Superseded by DRAFT, which keeps the minimal layer a solo practitioner can act on.

Links

💼 LinkedIn

🎓 42 Nice

🧪 Sayca Labs

Pinned Loading

  1. DRAFT DRAFT Public

    DRAFT is a bundle with its matrix, methods and tools, that keeps any project from forgetting why it was built the way it was. DRAFT items helps to tracks ideas, rules, design, build and field feedb…

    Makefile 3

  2. 42_Projects_unlicensed 42_Projects_unlicensed Public

    Projets 42 Nice. C/Unix, cursus RNCP-7. Code source délicencié, indexé par domaine.

    1

  3. RecursiveTriad RecursiveTriad Public

    [DEPRECATED] Un framework de cognition symbiotique développé depuis 2019. Employable depuis 2022 avec des LLM grand public. Structure l'intention humaine pour piloter l'exécution IA par consensus b…

    1