Skip to content
View giovanniromero-dev's full-sized avatar

Block or report giovanniromero-dev

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
giovanniromero-dev/README.md
+==================================================================================+
|                                                                                  |
|   G I O V A N N I   R O M E R O                                                  |
|                                                                                  |
|   FULL STACK DEVELOPER  //  AI ENGINEER                                          |
|                                                                                  |
|   LANGGRAPH | MCP | LANGCHAIN | RAG | AI AGENTS | AUTOMATION | FULL-STACK AI      |
|                                                                                  |
+==================================================================================+
giovanniromero@dev:~$ whoami
Full Stack Developer & AI Engineer · Building AI Agents & Agentic Systems ·
Cybersecurity Enthusiast.

giovanniromero@dev:~$ current_focus
> LangGraph state machines
> MCP servers and tool integrations
> LangChain / LCEL workflows
> RAG and knowledge assistants
> Full-stack AI products with APIs, dashboards, auth, databases, and deployment
> Security-aware boundaries for tools, data, permissions, logs, and secrets

./agent_graph

flowchart LR
    U[User / Business Workflow] --> UI[Next.js Interface]
    UI --> API[Backend API]
    API --> LG[LangGraph Orchestrator]

    LG --> PLAN[Planner Node]
    PLAN --> TOOLS[Tool Router]
    TOOLS --> MCP[MCP Server]
    MCP --> EXT[External APIs / Files / DB]

    LG --> RAG[RAG Retriever]
    RAG --> VDB[(Vector Store)]
    RAG --> DOCS[Knowledge Sources]

    LG --> REVIEW[Human Review]
    REVIEW --> OUT[Structured Output]
    OUT --> UI

    LG --> LOGS[Logs / Traces / Evaluations]
Loading

./mcp_manifest.json

{
  "profile": "Full Stack Developer & AI Engineer",
  "positioning": "Building AI Agents & Agentic Systems",
  "core_protocols": ["MCP", "LangGraph", "LangChain", "RAG"],
  "agent_patterns": [
    "tool calling",
    "stateful workflows",
    "human-in-the-loop review",
    "retrieval-augmented generation",
    "workflow automation",
    "structured outputs"
  ],
  "delivery": [
    "frontend interface",
    "backend API",
    "database model",
    "agent orchestration",
    "tool boundary design",
    "deployment notes"
  ],
  "security_mindset": [
    "auth boundaries",
    "least privilege tools",
    "secret handling",
    "audit logs",
    "safe API integration"
  ]
}

./build_matrix

Layer Tools What I Build
Agent Orchestration LangGraph, LangChain, LCEL Multi-step agents, state graphs, routers, retries, review loops
Tool Protocol MCP Tool schemas, MCP servers, connector boundaries, safe tool execution
Knowledge RAG, embeddings, vector stores Knowledge assistants, grounded answers, source-aware workflows
Product Next.js, React, TypeScript Interfaces, dashboards, admin panels, user workflows
Backend Node.js, Express, FastAPI APIs, integrations, auth flows, webhooks, automation services
Data MongoDB, PostgreSQL, SQL Persistence, run history, structured outputs, reporting
Automation Playwright, APIs, scheduled jobs Research agents, content workflows, business process automation
Security-Aware Engineering Auth, logs, secrets, validation Safer boundaries around data, users, tools, and external APIs

./services_vector

ai_agent_development             -> scoped agent workflows, tools, memory/state, handoff
mcp_server_tool_integration      -> MCP servers, tool schemas, permissions, API connectors
langgraph_workflow_builds        -> graph nodes, routing, retries, human review, observability
langchain_lcel_automation        -> chains, structured outputs, model/tool pipelines
rag_knowledge_assistants         -> ingestion, retrieval, grounded answers, source traceability
full_stack_ai_features           -> UI, API, database, auth, deployment, documentation
business_automation              -> repeatable workflows, integrations, dashboards, reports
security_aware_implementation    -> auth boundaries, secrets, logs, validation, safe APIs

./featured_work

giovanniromero@dev:~$ ls ~/repos --focus agentic-ai
Priority Repository Why it matters for this profile
01 osint-agent Research automation agent with Python, LangGraph, DeepSeek, Playwright, tool calling, public web data collection, and structured reporting workflows.
02 clinisight-ai AI medical insight prototype using FastAPI and MCP-style architecture for symptom extraction, PubMed retrieval, structured summaries, and API-driven workflows.
03 langgraph-foundations LangGraph foundations for AI agents: ReAct patterns, tool orchestration, state graphs, streaming, debugging, and agentic workflow control.
04 lcel-llm-translator LangChain LCEL and LangServe workflow for structured LLM pipelines, translation automation, API delivery, and full-stack AI integration patterns.
05 agentic-chatbot Agentic chatbot scaffold for LangGraph/LangChain tool orchestration, RAG retrieval, Tavily search, FAISS memory, and full-stack AI assistant experiments.
06 airplane-tracking-with-yolov8 Computer vision AI prototype for YOLOv8 object detection, tracking, video analysis, and visual intelligence workflows.

./agentic_system_design

1. define_scope
   - user goal
   - tool permissions
   - data boundaries
   - expected output

2. design_graph
   - nodes
   - edges
   - state
   - retry paths
   - review checkpoints

3. connect_tools
   - MCP server or direct API connector
   - schema validation
   - error handling
   - logs and traces

4. build_product_layer
   - UI
   - backend API
   - database
   - auth
   - admin controls

5. ship_iteration
   - test cases
   - usage docs
   - deployment notes
   - next-step roadmap

./learning_now

[COURSE] Complete MCP Bootcamp: Build Next-Gen AI Agents with MCP
[COURSE] Complete Agentic AI Bootcamp with LangGraph and LangChain
[FOCUS]  MCP server architecture, tool schemas, agent orchestration, RAG, evaluations
[GOAL]   Build useful AI systems that connect tools, data, APIs, and product interfaces

./contact

giovanniromero@dev:~$ ./connect --target professional
site      : https://giovanniromero.dev
linkedin  : https://www.linkedin.com/in/giovannideveloper/
x         : https://x.com/giovanni_dev_
email     : contact@giovanniromero.dev
giovanniromero@dev:~$ _

Pinned Loading

  1. osint-agent osint-agent Public

    Research automation agent with Python, LangGraph, DeepSeek, Playwright, tool calling, public web data collection, and structured reporting workflows.

    Python

  2. clinisight-ai clinisight-ai Public

    AI medical insight prototype using FastAPI and MCP-style architecture for symptom extraction, PubMed retrieval, structured summaries, and API-driven workflows.

    Jupyter Notebook

  3. langgraph-foundations langgraph-foundations Public

    LangGraph foundations for AI agents: ReAct patterns, tool orchestration, state graphs, streaming, debugging, and agentic workflow control.

    Jupyter Notebook

  4. lcel-llm-translator lcel-llm-translator Public

    LangChain LCEL and LangServe workflow for structured LLM pipelines, translation automation, API delivery, and full-stack AI integration patterns.

    Jupyter Notebook

  5. agentic-chatbot agentic-chatbot Public

    Agentic chatbot scaffold for LangGraph/LangChain tool orchestration, RAG retrieval, Tavily search, FAISS memory, and full-stack AI assistant experiments.

  6. airplane-tracking-with-yolov8 airplane-tracking-with-yolov8 Public

    Computer vision AI prototype for YOLOv8 object detection, tracking, video analysis, and visual intelligence workflows.

    Jupyter Notebook