Skip to content
View EZFRICA's full-sized avatar
🏠
Travailler à domicile
🏠
Travailler à domicile

Organizations

@gdgcloudabidjan

Block or report EZFRICA

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
EZFRICA/README.md

Hi 👋, I'm Ezekias BOKOVE

I'm a Google Developer Expert (GDE) for Google Cloud ☁︎.
I like to learn from others and share my knowledge.

👨🏻‍💻  Highlights

💡  Passionate about AI infrastructure, agentic systems, Cloud, Databases, APIs and DevOps. I enjoy contributing to projects that improve people's living conditions.
🔭  Currently deep-diving into LLM inference optimization, quantization, KV-cache management, serving frameworks (vLLM, TensorRT-LLM, llama.cpp), routing & batching.
🌍  Founder & ex-Lead Google Developer Groups GDG Cloud Abidjan
🌍  Co-Founder & Community Lead Google Developer Groups GDG Afrique Francophone
💬  Experienced public speaker on AI, Agentic AI, Google Cloud, Firebase, DevOps and PWA

🚀  Featured Projects

Context Lifecycle Engine (CLE): Framework for managing context switching in LLM-based agents, applying OS concepts (process control blocks, TLB, Markov prediction, Merkle DAG stores) to agent memory and context management.

Agent Processor Unit (APU): Open-source four-tier memory hierarchy for AI agents (in-process → Redis → Weaviate → Letta) with a bidirectional metadata routing algorithm, async priority-queue scheduler, and human-in-the-loop override dashboard. Also published as a Medium article series in the Google Cloud Community collection.

ALU Sage: Offline-first AI education platform for resource-constrained devices in Africa, built with Gemma, LanceDB on-device, and a LangGraph agent. Gemma Hackathon submission.

🛠  Tech Stack

Python  FastAPI  Docker  GitHub Actions  GitLab CI  Google Cloud  Firebase  PostgreSQL  MongoDB  Redis  Git  GitHub  GitLab  PyCharm  Visual Studio Code

🤝🏻  Connect with Me

⚙️  GitHub Analytics

Pinned Loading

  1. context-lifecycle-engine context-lifecycle-engine Public

    Detects recurring prompt patterns, compiles them into versioned agents, and automatically expires their proof of value when the underlying model drifts.

    Python

  2. alu-education-agent alu-education-agent Public

    A local-first Agentic Processor for Education (APU). Features a 3-tier DLL memory architecture (L1-L3), Socratic pedagogy powered by Gemini Flash, and cloud-synced course registries. Built with Lan…

    Python

  3. Agent-Processor-Unit Agent-Processor-Unit Public

    An Operating System Kernel for AI Agents. Implements an advanced MMU (Memory Management Unit) for persistent semantic storage, LRU paging, an asynchronous Semantic Scheduler and low-latency interac…

    Python

  4. google-cloud-mongodb-atlas-oidc google-cloud-mongodb-atlas-oidc Public

    A FastAPI test bench to validate end-to-end MongoDB Atlas Workload Identity Federation (OIDC) authentication from Google Cloud Run, with zero secrets stored.

    HTML

  5. ux-driven-agent-memory ux-driven-agent-memory Public

    A travel agent powered by a Living DLL + BMJ algorithm — where humans decide what the AI is allowed to know. Stack: Letta Cloud · Weaviate Cloud · Gemini · Streamlit

    Python

  6. ai-agent-memory ai-agent-memory Public

    Python