Skip to content
View CevenJKnowles's full-sized avatar
⭐
⭐
5 followers · 9 following

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx2Achievement: Pair ExtraordinaireAchievement: Quickdraw

Block or report CevenJKnowles

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.

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
CevenJKnowles/README.md

Ceven Jupiter Knowles

AI Systems Design · Governed LLM Infrastructure · Workflow Architecture

I build constraint-first LLM infrastructure: bounded execution, hard governance limits, architectural decisions that are documented before they are implemented.

The structural logic behind this work predates AI. Broadcasting, education, print, systems design — across each of these, the question has been the same: what allows a complex thing to hold together as conditions change? IO-III applies that same discipline to a new domain and a new scale.

The constraint is the feature.

IO-III Phase 10 complete. Nine phases hardened. Twenty-six Architecture Decision Records. The instrument is built.

Flagship

I0³ | Deterministic AI Runtime Architecture

Deterministic LLM control-plane runtime built in Python. Governance-first architecture with 26 Architecture Decision Records, bounded execution enforcement (hard limits on audit and revision passes), recursive content-safety logging, steward supervision mode, a multi-provider adapter layer, and an architectural invariant test suite that grows with each design decision. Ten complete phases. Phase 11 in planning. Latest stable: v1.0.0. MIT licence.

Infrastructure & Governance

cjk-knowledgehub-governance Internal control and governance layer for the KnowledgeHub system. Defines ADRs, phase freeze protocols, control gates, deterministic naming conventions, and structural invariants.

CJK KnowledgeHub Public portfolio and knowledge site built with MkDocs. Live at cevenknowles.com. Documents the structural method, project history, and operating infrastructure across domains.

Projects

Job Application Tracker Local desktop application built in Python with PyQt6, SQLite, Matplotlib, and a LaTeX export pipeline. Normalised relational database with user-editable reference tables and embedded analytics. Designed as a compact internal product system, not a spreadsheet. Phase 1 in progress.

Prompt Pattern Library Open-source taxonomy and atomic pattern library for structured prompt engineering. Taxonomy v1.0 locked: 19 domains covering instruction and control, workflow automation, evaluation, safety and ethics, failure modes, and meta-system control. Atomic pattern library in development.

Capabilities

  • AI Systems Design & Governed LLM Infrastructure
  • Workflow Architecture & Deterministic System Design
  • Architecture Decision Records & Documentation Governance
  • Bounded Execution Modelling & Runtime Safety
  • Responsible AI & Content Safety Architecture

Connect

Open to technical conversations, collaboration, and the right opportunities.

Pinned Loading

  1. io-iii-runtime io-iii-runtime Public

    Deterministic LLM control-plane runtime. Governance-first architecture with bounded execution, hard audit limits, and an architectural invariant test suite.

    Python 1

  2. cjk-knowledgehub-governance cjk-knowledgehub-governance Public

    Documentation for the creation and project management of the git repo "CJK_KnowledgeHub"

  3. cjk-knowledgehub cjk-knowledgehub Public

    Portfolio bridging media design background into current AI/tech work. Showcases interdisciplinary projects, skill examples, and professional trajectory.

    HTML

  4. prompt-pattern-library prompt-pattern-library Public

    Developing a comprehensive taxonomy system for prompt patterns (18 domains, 150+ categories). Includes definitions, use cases, error handling, and inter-pattern compatibility. Active work in progress.

    1