Skip to content
View Adikasz's full-sized avatar
1 follower · 0 following
  • Hungary

Achievements

Achievement: Pair ExtraordinaireAchievement: YOLO

Achievements

Achievement: Pair ExtraordinaireAchievement: YOLO

Block or report Adikasz

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
Adikasz/README.md

Hi, I'm Ádám Nagy ⚡

AI Product Engineer | Founder @ PlanSmart

I build autonomous, production-grade AI systems that solve expensive B2B bottlenecks. My background bridges the gap between raw LLM capabilities and measurable business ROI. I don't just write prompts; I engineer deterministic, zero-error agentic workflows using the WAT (Workflows, Agents, Tools) architecture and the Model Context Protocol (MCP).

🧠 The Hybrid Advantage

As an Economics graduate (BGE) turned AI Architect, I approach software engineering through a product management lens. I focus on:

  • Asymmetric Intelligence: Building agents that give sales and strategy teams an unfair advantage.
  • Zero-Error Engineering: Enforcing strict typing, 100% offline-testable CI/CD pipelines, and mathematical isolation of untrusted data.
  • Enterprise Extensibility: Abstracting all external integrations behind official MCP (FastMCP) servers to maintain clean reasoning loops.

🚀 Featured Flagship Project

Scout: Autonomous B2B Intelligence Engine
An enterprise-grade competitive intelligence agent. It ingests a target company and a strategic context, autonomously browses the live web (Tavily MCP), synthesizes the data via the Claude API, and outputs a CTO-ready Executive Briefing.

  • Tech: Python 3.12+, Anthropic Messages API, FastMCP, Pydantic, Pytest (AST-guarded network blocking).

🛠️ Tech Stack & Architecture

  • AI/LLMs: Anthropic Claude API (Sonnet/Opus), Model Context Protocol (MCP), Agentic Workflows (WAT)
  • Backend: Python 3.12+, Asyncio, Pydantic (Strict configuration boundaries)
  • Engineering Quality: Pytest, Ruff (Lint/Format), Mypy (--strict), GitHub Actions
  • Integrations: Tavily (Live Web Intelligence), FastMCP

🤝 Let's Connect

Looking to automate complex workflows or build custom intelligence agents for your enterprise? Let's talk.

"Business-first automation, engineered to production standards."

Pinned Loading

  1. scout-intelligence-agent scout-intelligence-agent Public

    Autonomous B2B Research & Content Engine powered by Claude Sonnet and MCP.

    Python

  2. ppc-monitor ppc-monitor Public

    Autonomous async AI orchestration system for large-scale AdOps monitoring.

    Python

  3. Enterprise-RAG-Knowledge-Engine Enterprise-RAG-Knowledge-Engine Public

    Enterprise-grade RAG and Agentic Framework powering PlanSmart Agency. Features strict SSRF defense, FinOps cost-optimization, and Claude orchestration via MCP.

    Python