2026-09-23·by Sijie Wang#standmeet#product

introduction

Product introduction

Parent: product · the why is in motivation

What it is

StandMeet inverts the compression. You curate a knowledge base of who you are and what you've actually done; then you — or a visitor — query it, and an AI answers in your voice, grounded in your own words, with citations. It's the first queryable, network-native self-presentation system — your personal truth layer, what an AI consults when someone asks "should I trust this person, and are they a fit?"

The corpus is yours to curate and push (via MCP, or by syncing your own note vault), never scraped. The act of attaching it is the trust boundary; nothing surfaces that you didn't put in. Both routes keep that boundary in the same place — the vault sync simply moves the moment of choosing from pushing to writing it down at all (thesis).

The first user is you — an external brain over your own thinking. Others querying it is a second surface, and hiring is only one application of that, not the headline (see boundaries).

Modes are roles

Not one app with four access levels — four apps sharing one corpus, framed by who's asking:

  • self — your retrieval & drafting surface
  • public — open Q&A for anyone
  • recruiter — evaluation, with the agent pre-screening them against your criteria
  • friend — casual, lower-formality exploration

Roles sit on the access layer: ACL is the hard wall on what's reachable; the role is the soft framing on what gets surfaced.

Owner side

Retrieval ("what did I think about X?") · draft in your voice (kills the core pain: not wanting to repeat what you've already said) · articulation aid · cross-corpus synthesis · curation & visibility suggestions.

Visitor side

Query you in your voice (the core) · meeting request with the agent pre-screening the asker against your filters (tech-driven team, real IC ladder, right stage and size) · compatibility / JD check · casual exploration.

Across both sides: answers carry citations, private topics stay private, and with a calendar connected a visitor can book a follow-up inside the conversation. Owners invite others via access codes (roles + quotas), publish long-form writings, and extend the agent with skills / MCP.

Who it's for

People whose first impression is high-information and high-stakes, tired of being averaged out — builders, founders, researchers, the kind of person who thinks but won't perform on LinkedIn — and the recruiters, investors, and collaborators sizing them up. It serves networked thinkers, not job-seekers.

Stance

Anti-résumé, anti-performance, anti-AI-slop. Surviving an agent's compression means being specific and opinionated, so StandMeet carries the differentiated, provable detail that emotional marketing throws away. No MVP-grade slop; the craftsmanship version. The surface is typographic and restrained — Are.na / Stripe Press / Readwise Reader, not a chat app: no alternating bubbles, no hype.

about this entry

One of sijie's wiki entries. The AI on this site is grounded in the same corpus and answers in sijie's voice, with citations back to entries like this one — answering costs sijie money, so it waits behind a code: enter an access code →