Candidate experiment · AI Solutions Manager

AI Recruiter
vs. Me.

I thought writing “passionate about AI” was a little boring. So I built a system that can interview my profile, challenge it, and show how I actually think about AI-enabled business problems.

LOADING · evidence base
This prototype is designed to answer a more useful question than a static CV: “Why this candidate, why not him, and is he worth interviewing?”
💬 Ask my AI directlygrounded on my profile · no web search
Each new question uses at most one fast-model call; identical questions are cached.
↓ Pick an action — the result opens immediately below.

LIVE OUTPUT

waiting for input
Ask a question or choose an action above. The answer appears here automatically. Live interactions use one structured LLM call at most; identical repeats are cached.

GROUNDING

loading context
Loading the local evidence pack…
1 maxAPI call per action
Cacheidentical repeats
JSON Schemastrict output
Dockerdeployment
Prototype note: The profile is deliberately positioned like a strong recruiter pitch: transferable evidence is made explicit, working-style statements are identified as self-description, and missing credentials are not invented. Live AI uses the same frozen evidence pack, strict Structured Outputs, database cache and explicit daily call limits.