Ask an agent about me.

This site runs a Model Context Protocol server. Point any MCP client at it and your assistant can answer questions about my experience, stack and projects from the same source that renders my CV — no scraping, no stale copies.

Connect

Streamable HTTP, no authentication, no session to keep. Add this to your client config:

{
  "mcpServers": {
    "snowye": {
      "type": "http",
      "url": "https://snowye.dev/api/mcp"
    }
  }
}

Autodiscovery

Clients that only know the domain can find the server and its tools here:

• /.well-known/mcp.json
• /api/mcp (GET)
• /llms.txt

Tools

Every tool is a pure function of my résumé source, so nothing here can drift from the CV page.

get_profile

Who Gabriel Trzimajewski is: name, title, location, contact, summary, and whether he is currently open to roles.

get_resume

The complete CV as Markdown: roles, dates, achievements, education, skills, languages.

get_experience

Structured work history, most recent first, with employer, position, dates and highlights.

get_skills

Technologies: the stack he is applying with, his signature skills, the full known-for set, and the grouped skills from the CV.

get_projects

Side projects and open source work, newest year first, with links and descriptions.

search_resume

Full-text search across the CV. Returns the matching sentences with the role they belong to — use it to answer "has he worked with X?".

get_contact

How to reach Gabriel: email, social profiles, and his booking link.

Try it

Or call it directly over curl:

curl -X POST https://snowye.dev/api/mcp \
  -H 'content-type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
       "params":{"name":"search_resume",
                 "arguments":{"query":"kafka"}}}'

Not using MCP?

The same facts are served as plain text and Markdown: llms.txt, cv.md, and the CV page with schema.org markup.