Nice job. I've seen a couple of these on HN (Resume-context driven chat interfaces). [1] [2]<p>I can't speak to whether this could become a realistic standard but you might want to reach out to @jhgaylor who took a stab at trying to build an MCP server around this concept. [3]<p>[1] - <a href="https://www.jon-olson.com/resume_ai" rel="nofollow">https://www.jon-olson.com/resume_ai</a><p>[2] - <a href="https://replicant.im/alex" rel="nofollow">https://replicant.im/alex</a><p>[3] - <a href="https://news.ycombinator.com/item?id=43891245">https://news.ycombinator.com/item?id=43891245</a>

Hey HN! I'm the builder here.<p>If you're interested in the full story and more details, I also wrote about this on LinkedIn: <a href="https://www.linkedin.com/posts/charlie-tianle-cheng-6147a4325_opentowork-newgrad2026-activity-7424451805322940417-H2wV" rel="nofollow">https://www.linkedin.com/posts/charlie-tianle-cheng-6147a432...</a><p>The technical implementation treated the iteration process like training a model: test the AI's responses, measure the "loss" against what I wanted, backpropagate by adjusting prompts/RAG/CRUD, and repeat.<p>Happy to answer any questions about the tech stack, the AI architecture, or the broader vision!

Wow.<p>> Build a platform where anyone can create their AI twin for genuine matching.<p>Add a premium tier that deepfakes you into each opening's "AI"-researched ideal candidate.<p>And a super premium tier to deliver the exclusive best fake for each particular opening.<p>Once you get traction, offer recruiters a filter that removes the fakes for $$$, but instead deliver just improved fakery.<p>Let recruiters pay $$$$$ to have their competitors get only fakes.<p>But be quick, else be beaten to it by Cory Doctorow or Charlie Brooker :)