Skip to content
View Tony-Stark2025's full-sized avatar
6 followers · 1 following

Highlights

  • Pro

Block or report Tony-Stark2025

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
Tony-Stark2025/README.md

Bright Onwe

AI Systems & Autonomous Agentic Engineer

Specializing in Multi-Agent Architectures, Model Context Protocol (MCP), and LLM Observability on Google Cloud

Google Developer Profile Kaggle Profile Twitter/X Email

🚀 About Me

I am an AI Systems Engineer and Electrical Engineering student at the University of Ibadan, building resilient, production-ready autonomous multi-agent systems, Model Context Protocol (MCP) tool ecosystems, and enterprise agent observability platforms.

  • 🧠 Autonomous Agentic Systems: Multi-agent asynchronous DAGs, human-in-the-loop, guardrails, and deterministic tool bindings.
  • ⚡ Model Context Protocol (MCP): Authoring stdio & SSE MCP servers/clients interfacing real-time observability telemetry (Grafana, GitHub, Jira, Slack).
  • ☁️ Google Cloud & Vertex AI: Vertex AI Reasoning Engines, Cloud Run deployments, GCP IAM, ..d.
  • 🏆 Honors & Recognition: Google Developer Diamond League status (6,000+ points) across specialized AI agent tracks.

🌟 Flagship Projects & Case Studies

🔥 Prometheus — Enterprise Workstream Observability Platform

Status Gemini MCP

Autonomous AI Chief of Staff correlating cross-functional engineering telemetry (GitHub PRs, Jira epics, Slack discussions) via ABAC security perimeters to proactively diagnose delivery blockers.

  • Key Highlights:
    • 🌐 Live Production Deployments: Cloud Run web application & Vertex AI Agent Engine (us-central1).
    • 🔌 Hybrid MCP Protocol: Dual stdio and Server-Sent Events (SSE) MCP servers.
    • 🛡️ Enterprise Security & ABAC: Fine-grained Attribute-Based Access Control ($P(U, R)$) and inline PII sanitization.
    • 🧪 Hermetic Quality: Complete 5-tier test matrix with 157 hermetic automated tests.
  • Tech Stack: Python 3.11, FastAPI, Gemini 3.7 Flash, Google Vertex AI, Model Context Protocol, OpenTelemetry, Docker, Cloud Run.
  • Live Demo: Prometheus Live Dashboard

🎬 Showrunner — Autonomous Studio Operations Copilot

Live Demo Vertex AI Grafana MCP

Enterprise VFX & Virtual Production Ops Copilot built for the Agentic Cinema Hackathon (Grafana Labs Track). Autonomously monitors, diagnoses, and self-heals VFX 3D render farms (Blender Cycles, Unreal Engine 5.4 Nanite) and virtual production LED volumes.

  • Key Highlights:
    • 🤖 Specialized Multi-Agent Crew: Sentinel, Diagnostic, Remediation, and Executive agents with role-tailored reasoning budgets.
    • 📊 Grafana MCP Integration: Direct tool querying for Loki logs, Tempo distributed traces, and Mimir telemetry.
    • 🖥️ Cinematic Command Center: Dark-mode high-density operational UI with real-time incident simulators.
  • Tech Stack: TypeScript, Next.js 14, Vertex AI (Gemini 3.7 Flash), Grafana MCP, TailwindCSS, Google Cloud Run.
  • Live Demo: Showrunner Production URL

🛠️ Agent DevKit — Automated Prompt Engineering & Synthetic Evaluator

Command-line tool automating prompt generation, synthetic evaluation, and iterative refinement using Google's Agent Development Kit and Gemini models.

  • Tech Stack: Python, Google Gemini SDK, CLI System Design, Evaluation Frameworks.

🥗 PlatePledge — AI-Powered Food Waste Mitigation Platform

Real-time surplus food redistribution system matching local restaurants and event donors with food banks and shelters using Gemini multimodal classification and geolocation routing.

  • Tech Stack: Angular 20, TypeScript, Google GenAI SDK, TailwindCSS.

🛠️ Technical Skills & Ecosystem

Domain Technologies & Frameworks
Agentic AI & LLMs Model Context Protocol (MCP), Multi-Agent DAG Orchestration, Google ADK, Vertex AI Agent Engine, Gemini 3.7 Flash Hybrid Reasoning, Tool Calling
Cloud & Deployment Google Cloud Platform (Cloud Run, Vertex AI, Cloud Build, Secret Manager, IAM, ADC), Docker
Backend & APIs Python (FastAPI, AsyncIO, Pydantic, Uvicorn), Node.js, Express, REST APIs, Server-Sent Events (SSE)
Frontend & UI TypeScript, React, Next.js, Angular, Tailwind CSS, Vite
Observability & Testing OpenTelemetry, Grafana (Loki, Tempo, Mimir), pytest (Hermetic Testing Suites), LogQL
Databases & State Firestore, SQLite Checkpointing, LocalStorage / State Machines

📬 Connect & Collaborate

I am open to AI Engineering roles, contract / freelance multi-agent systems consulting, and innovative hackathon collaborations.

Popular repositories Loading

  1. PlatePledge PlatePledge Public

    AI-powered surplus food redistribution platform connecting donors with local shelters in real-time

    TypeScript 5

  2. Tony-Stark2025 Tony-Stark2025 Public

    Config files for my GitHub profile.

  3. Agent_DevKit Agent_DevKit Public

    Automated prompt evaluation, generation, and iterative refinement CLI tool built with Google Gemini SDK

    Python

  4. A2A_ADK_MCP A2A_ADK_MCP Public

    Forked from Roji-val/A2A_ADK_MCP

    Multi-Agent Systems with Google's Agent Development Kit + A2A + MCP

    Jupyter Notebook

  5. opik opik Public

    Forked from comet-ml/opik

    Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.

    Python

  6. Dolphin Dolphin Public

    Forked from bytedance/Dolphin

    The official repo for “Dolphin: Document Image Parsing via Heterogeneous Anchor Prompting”, ACL, 2025.

    Python