Lan Xuezhao

Who they are

Lan Xuezhao is Founder and Managing Partner of Basis Set Ventures — a Ph.D. cognitive scientist who built Pascal, a proprietary AI system for sourcing startups, before most VCs were talking about AI tooling.

Person

Lan founded Basis Set Ventures in 2017, launching with over $100M in initial funding — unusually capitalized from day one for an early-stage AI-focused firm. The arc before that is distinctive: a Ph.D. in Psychology and Quantitative Cognitive Science from the University of Michigan (2009), an M.A. in Statistics from the same institution (2008), then a hard pivot into consulting at McKinsey as an Engagement Manager, followed by Corporate Development Strategy (M&A) at Dropbox. Before VC, she also founded a brain-training games startup for children that was later acquired. The through-line is quantitative rigor applied to messy human systems — cognitive science to M&A to venture. She built two proprietary internal tools at Basis Set: Pascal, an AI system for sourcing startups, and HyperGrowth, a platform for connecting portfolio companies with mentors. Possibly — her cognitive science background shapes how she thinks about founder assessment and AI's effect on knowledge work. She writes on LinkedIn about early-stage AI investing, AI-enhanced learning, and hardware reindustrialization.

Company

Basis Set Ventures closed its $250 million Fund IV in January 2026, its largest vehicle to date, positioned explicitly as capital for partnering with founders at the earliest stage. The fund close comes as AI deal volume climbs and Basis Set's portfolio includes notable names — Scale AI, Drata, Quince, CuspAI, and Mem0. A recent partnership announcement with Mem0 signals continued focus on AI memory and agent infrastructure. The firm runs proprietary tools — Pascal for deal sourcing and HyperGrowth for portfolio support — that are part of its differentiation pitch to founders.

Market

Early-stage AI venture is intensely competitive in 2026, with generalist firms and specialist AI funds both chasing seed-stage deals. Basis Set sits in the specialist camp, focused on AI and automation startups, competing for allocation against firms with larger platforms and bigger brand. AI startup funding has surged, which inflates entry valuations at seed — a headwind for any fund writing early checks at disciplined prices.

Network

Lan publicly thanked Connie Loizos, Editor-in-Chief, in a LinkedIn post — suggesting a media relationship worth noting. Her portfolio engagement spans Scale AI, Drata, Quince, CuspAI, and Mem0, where she holds board or observer seats as lead seed investor.

  • Connie Loizos· Editor-in-Chief (media contact, publicly acknowledged)

How they likely show up

  • Long tenure at Basis Set since 2017 (nearly nine years, fund I through Fund IV) → she thinks in fund cycles, not quarterly cycles, and likely evaluates relationships over years.
  • Ph.D. in Quantitative Cognitive Science and M.A. in Statistics → she will interrogate the numbers behind a thesis; expect her to probe assumptions quantitatively.
  • Built Pascal (AI sourcing) and HyperGrowth (portfolio mentoring) as internal tools → she builds systems to scale judgment, not just gut-driven dealmaking.
  • Career moved from academia → McKinsey → Dropbox M&A → founding a fund → she's comfortable in structured environments but left each one to own more; agency matters to her.
  • Possibly — occasional LinkedIn posts on AI-enhanced learning and hardware reindustrialization suggest she's thinking beyond pure software, which may shape where Fund IV deploys.
  • Founded a brain-training games startup before pivoting to VC → she's been an operator, not just an observer; she likely extends that credibility explicitly when talking to founders.

Conversation tips

  • → Reference Pascal or HyperGrowth specifically — she built proprietary tooling before it was fashionable in VC, and it's a genuine point of differentiation she'll want to talk about.
  • → Ask about the cognitive science background and how it shapes her founder or market assessment — it's an unusual path into VC and she almost certainly has a developed point of view on it.
  • → Come with a specific thesis, not a general AI enthusiasm — she writes about early-stage AI investing with real specificity and will disengage from surface-level AI hype.
  • → If you have a hardware or reindustrialization angle, surface it — her public writing suggests she's thinking beyond software AI, which is less common in SF seed funds.

Toolbox

Openers

  • Open on Pascal — she built an AI system for startup sourcing inside Basis Set years before most VCs were doing this; asking how it's evolved with the current generation of models is a natural and genuine entry point.
  • Lead with the Fund IV close in January 2026 — $250 million, the firm's largest, explicitly positioned for the earliest stage; ask what the thesis shift looks like from prior funds given how the AI market has matured.
  • Reference the Mem0 partnership announcement — it signals where she's placing bets on AI agent infrastructure, and it's recent enough to be a live conversation, not a retrospective one.

Discovery questions

  1. You built Pascal to source deals algorithmically — how much of Basis Set's conviction on a company now comes from Pascal versus the human judgment layer on top of it?
  2. Fund IV is $250 million targeting the earliest stage in what is now a very crowded AI seed market — how has the entry-price discipline changed, if at all, when every round is oversubscribed?
  3. Your background runs from cognitive science to McKinsey M&A to founding a VC — when you're evaluating a founder, which of those lenses do you actually reach for first?

Avoid

Don't pitch AI in generic terms or lead with category-level enthusiasm — she has written specifically about AI investing, built her own AI tooling, and will expect precision about what a company actually does differently.

Make it yours

Tailor these openers to what you sell

These openers are generic. Sign in and tell Brief what you sell — it rewrites the hooks and questions around your pitch.

Sources

Brief on your next meeting?

Type any name. Get a structured pre-meeting brief in seconds.

Try Brief →

Generated by briefthecall.com from public web sources on June 5, 2026. Each claim is linked to its source above.

Automatically generated by AI from public sources. May be inaccurate or out of date. Remove or correct this profile →