• AI agents — tool calling, orchestration, multi-step workflows
• AI integrations — MCP servers, OAuth, REST APIs, external services
• Backend systems — Python, FastAPI, async services, WebSockets
• Automation — turning repetitive business processes into reliable workflows
• Production systems — Docker, Redis, PostgreSQL, telemetry, CI/CD
• AI security — securing agent tools, MCP servers and AI-powered workflows

I work end-to-end: architecture → implementation → integration → testing → deployment → debugging.

MCP server that connects an AI assistant to LinkedIn's official API for publishing and scheduling posts.

Python · MCP · OAuth · REST API

Security scanner for MCP servers. Inspects tools, prompts and capabilities for poisoning, injection surfaces and excessive permissions.

Python · MCP · Security automation

AI-assisted security automation platform with agent orchestration, scope controls, reporting and a concurrent Go core.

Python · Go · MCP · Agents

Automation project for researching and evaluating YouTube content opportunities.

Python · Automation · Data analysis

Example AI-agent architecture for analysing sales performance using CRM and telephony data.

Python · APIs · AI agents · Architecture

A large private system I build end-to-end: event-driven backend, AI decision layer, external data integrations, risk controls, automation, telemetry and operational dashboard.

Python · FastAPI · asyncio · WebSocket · Redis · PostgreSQL · Docker · React

The repository is private, but I can demonstrate the architecture and implementation during an interview.

I prefer systems where AI is connected to real tools and measurable workflows rather than isolated chat interfaces.

Typical flow:

Request → Agent → Tools/APIs → Data → Reasoning → Action → Verification → Observability

Key principles:

• deterministic logic where possible;
• AI where reasoning adds value;
• explicit tool boundaries and permissions;
• human approval for high-impact actions;
• logs, metrics and traceable results;
• reproducible deployments.

Languages: Python, Go, JavaScript/TypeScript, SQL, Bash

Backend: FastAPI, asyncio, WebSockets, REST, Pydantic

Data & infrastructure: PostgreSQL, Redis, Docker, Linux, Nginx

AI: AI agents, MCP, tool calling, orchestration, LLM integrations, ML pipelines

Engineering: Git, GitHub Actions, testing, observability, API integrations, OAuth