Leigh Marie Braswell

Snapshot

• Role: Partner @ Kleiner Perkins (joined May 2023). Focus: infrastructure & ML-application founders.
• Focus: Self-described “relatively generalist” but in practice AI infra + applied ML; builds “unfair advantages” via her technical network. Backed some of the buzziest 2025 AI deals (Windsurf, Neon).
• Location: San Francisco.
• The 1-liner: An ex-Scale-AI engineer-turned-VC with a competition-math/poker brain and a hard-nosed “bullsh*t detector for AI companies” — widely flagged (Fortune) as a rising star in VC despite only ~6 years in the seat.
• Handles: LinkedIn /in/leigh-marie-braswell. X/Twitter: @LM_Braswell (~20.8K followers; bio “partner @kleinerperkins; previously investing @foundersfund, first pm & early eng @scale_ai; if you like infra, ML, or poker let’s chat”). Substack: leighmariebraswell.substack.com. ✅ confirmed 2026-06-01.

Background

Career arc (corroborated across KP welcome post, Fortune poker profile, theorg.com, and her own MLOps-podcast account):

• Phillips Exeter Academy — Valedictorian. ^[single-source: theorg.com]
• MIT — Math with Computer Science (humanities concentration in Economics). Top-10 competitive female mathematician in the US in high school; came up through math olympiads. Met Scale AI’s Alexandr Wang first at a high-school math competition (junior year), reconnected at the MIT Poker Club.
• Jane Street — Assistant Trader / intern (quant trading). ← This is where she learned poker (“playing every day against people at Jane Street”). ⚠️ CRM correction below — this, not Citadel/Two Sigma, is the real quant stop.
• Blend — software development (early role). ^[single-source: her MLOps-podcast self-account]
• Google — Associate Product Manager / ML role.
• Scale AI — early engineer (one of the first ~ten, joined as an intern at ~3 people) → first Product Manager. Built and led the LiDAR / 3D sensor-fusion (LiDAR+RADAR+camera) annotation products for AVs/robotics/AR-VR. Multiple patents. Stayed through ~2020; “Back then, Scale was the API for human labor.”
• Angel investing — began 2020 (her last year at Scale). First check: Ambience Healthcare (2020), now ~$1.1B. Other angel bets: Modal, Monad, Succinct Labs, Ambience, Alquist Robotics, Opal Camera, etc.
• Founders Fund — Principal, ~2020/21–2023 (~2+ yrs), data & ML infra/apps. (FF deal: Chronosphere → acquired by Palo Alto Networks.)
• Kleiner Perkins — Partner, May 2023 – present. Joined “at the firm’s turnaround” (Hamid/Fushman era).

⚠️ CRM CORRECTION (high priority): the CRM seed lists Leigh as “ex-engineer Citadel & Two Sigma.” This is wrong. Across 4+ independent sources (KP “Welcoming…” post, Fortune poker profile, theorg.com org-chart, and her own words on the MLOps Coffee Sessions podcast) her quant-trading stop is Jane Street (assistant trader/intern) — the firm where she learned poker. No source places her at Citadel or Two Sigma. Likely a quant-firm name-swap in the CRM. Do not reference Citadel/Two Sigma to her in person.

Investing

• Thesis / focus: Infrastructure + ML applications; “everything is AI now,” but she keeps a practitioner’s skepticism. Credits Scale for a “solid ‘bullsh*t detector’ for AI companies.” Doesn’t “exactly consider herself an AI investor” despite the portfolio.
• Notable investments / board roles:
  • KP: Avoca (led Series A), Windsurf (board observer/director; early angel → KP), Neon (→ Databricks), Nooks, Convoke (co-led $8.6M seed w/ Dimension), Reevo, Forge, Persona (board), Exafunction, Arena AI, Grit.
  • Founders Fund: Chronosphere (→ Palo Alto Networks), Persona, Neon.
• Style: Leads and co-leads early-stage. Network-as-moat sourcing — runs weekly poker nights in SF (with Wang + top tech figures) that double as deal sourcing, hiring, and relationship building. Verticalize first: “be the best in that one area… and then expand outwards” (cites Scale’s AV-labeling wedge as the model).
• On metrics / diligence: Sharp on revenue-quality — “ARR can mean wildly, wildly different things… you can’t just take revenue at face value,” especially when AI usage costs are high and “some firms are losing money the more that users use them.” On PMF: “If you’re not feeling like everything is on fire all the time, there’s probably not product-market fit.”

Public voice (EXHAUSTIVE)

The load-bearing arc for init.inc: her evolving view on AI agents.

• Dec 2024 (Fortune Brainstorm AI panel): “agents don’t work… They do not yet work reliably for the vast majority of use cases.” Caveat: “Yes, agents can really speed you up,” but need a human in the loop “for basically all of the use cases.” Analogy: self-driving cars still need human guidance.
• Apr 2026 (Avoca Series A, Fortune): she now leads KP into an AI-agents-for-services company. Her resolution of the 2024 skepticism is vertical scope + integration depth: “What excites me about Avoca is that they’re applying AI where execution is the real bottleneck… something that feels less like a tool and more like core infrastructure for how this industry operates.” On the market: “This is definitely an industry that’s been overlooked by Silicon Valley… people don’t realize how big it is.” “A company like Avoca is a necessary bridge between Silicon Valley and Main Street.” On first-mover claims: “I like the phrase ‘head start’ better. It’s more intellectually honest.”

Other public material:

• Sourcery.vc — long-form interview (“Kleiner Perkins 10B+ AUM & The Rise of AI"). KP fund math (KP21 825M, KP Select III $1.2B,$2B+ Rise-of-AI, ~9 investors); Scale stories (“the Alex API,” helping OpenAI label GPT-2 — “what even is this?”; the “talent vortex” concept). On code-gen: a “war zone” (Copilot, Devin, Cursor); “the moat… is just like how fast they can move,” currently “speed is kind of the only moat.” On Windsurf: “There’s no limit to their ambition. They don’t just want to make 10x developers better — they want to make everyone a developer.” The “seed-strapped” thesis: tiny teams with 10x revenue growth that “don’t need to raise a Series A… You may see companies go from a seed to a Series C. It’s just gonna be weird.”
• Fortune poker profile (Feb 2026): “If the odds are in your favor, you push your chips to the center.” “We didn’t go out and party in New York — we stayed in and played poker.” “I barely knew what VC as a concept was.”
• MLOps Coffee Sessions #81 (MLOps.community, w/ Davis Treybig, ~Jun 2022) + Software Engineering Daily (Oct 2021, “Modern Data Infrastructure and Tools”) + Featureform podcast (“MLOps From A Practitioner Turned VC”) — her MLOps/infra thesis: lack of standards, build-vs-buy friction, verticalize-then-expand, clear-value-prop bar vs SageMaker.
• Video (YouTube — queued for transcription, transcript_queue.txt):
  • “Leigh Marie Braswell on the Future of Generative AI” — KP @ AI House Davos panel, Apr 15 2025 (gk7h4ARnZEY) — gen-AI vs hyperscalers.
  • “Windsurf: The Making of a Billion-Dollar AI Company” — w/ Molly O’Shea, May 2 2025 (pm6_qNp21w0).
  • “#217 Varun Mohan (Codeium/Windsurf) w/ Leigh Marie Braswell: Limitless” — Nov 18 2024 (EKRghiBVf_o) — code-gen, her Windsurf bet.
• KP “Welcoming Leigh Marie Braswell” post — KP on her: “intellect, technical depth, sharp wit, and unbelievable hustle”; founders’ “first call for advice.”
• Social: LinkedIn active; X/Twitter handle unconfirmed ^[ambiguous].

Personal & interests (rapport)

• From rural Alabama (KP welcome post + Fortune). Strong “unlikely-VC-from-Alabama” identity.
• Competition math — math olympiads, top-10 female mathematician in the US in HS. Phillips Exeter valedictorian → MIT.
• Poker — serious, daily-practice level; learned at Jane Street; hosts weekly SF poker nights (with Alexandr Wang + top tech). Poker is both a hobby and her sourcing/relationship engine. ← highest-value rapport hook.
• Endurance / outdoors — KP bio: “loves to play poker, run long distances, and scuba dive.” (Long-distance running + scuba diving.)
• Voice/temperament: intellectually intense, competitive, intellectually-honest-to-a-fault (“head start” not “first-mover”), practitioner’s skepticism, dry sharp wit. Respects rigor and non-consensus; allergic to hype/“ARR theater.”
• Network: tight with Alexandr Wang (Scale) — a recurring node in her sourcing and a possible mutual-connection vector.

Fundraise Relationship (init.inc)

• Relationship facts: Leigh is Init Intelligence’s internal champion / sponsor at KP — not the active POC (she handed Sazzad to Nadia); she is the sponsor edge behind the live Nadia thread. Her thesis is a strong fit: she already led Avoca’s Series A on the “vertical AI that absorbs operational labor where execution is the bottleneck” logic. She is on record (Dec 2024) that “agents don’t work… not reliably,” and resolved it (Apr 2026 Avoca) via vertical scope + integration depth; she dislikes “ARR theater.”
• Conversation log (newest first):
  • 2026-06-01 — Met Leigh + Nadia (Zoom “Sazzad & KP,” 12:00pm PT). Went well — both “big fans” (“incredible background”); had read the memo + profile. Leigh ran the diligence Qs (outcome-ownership: in-house vs MSP-acquisition; hiring; GTM/ICP; round = $5M). Leigh defined the path: KP shares notes w/ the broader team → if green-lit, “one or at most two more meetings before a final decision.” Leigh is freshly back from maternity leave. Sazzad set a Fri 6/5 offer deadline + offered to help expedite. (Notes: _raw/meeting-prep/2026-06-01-kp-meeting-NOTES.md.)
  • 2026-05-05 — Leigh connected Sazzad internally to Nadia (warm handoff). Her relative quiet since is expected (handed off) — not a cold signal.
  • 2026-05-05 — Met Leigh — went well (positive read).
  • 2026-05-04 — Kushal made the intro.
• Mutual connections / warm path: Kushal (intro’d her). She is ex-Founders Fund (FF-alumni network). Alexandr Wang / Scale network is a possible secondary vector.
• Personal & rapport notes (about this person): She rewards rigor and non-consensus thinking. Genuine surfaces: poker / expected-value framing (“odds in your favor, push chips to the center”), competition math, the Alabama-to-MIT story, endurance running / scuba. She has a practitioner’s “bullsh*t detector” and is sharp on revenue quality / unit economics.
• Live stage: tracked in the CRM (Google Sheet) — not duplicated here.

2026-06-01 refresh (net-new, dual-engine verified)

• Roadrunner — KP’s first incubation since Glean (announced May 12, 2026). AI-native CPQ / “quote-to-cash” (“PQA — prompt, quote, approve”). $27M\ast \ast total:\ast \ast$5.2M seed led by Mamoon Hamid (KP) + $22M Series A led by Trae Stephens (Founders Fund). CEO Joubin Mirzadegan (KP partner / Grit host). Leigh is the active internal cheerleader (recruiting publicly: “an absurdly strong team… real customer demand from day 1”). Signal: KP currently bets on rebuilding entrenched enterprise software AI-natively with day-one demand.
• Alkira → Lumen, $475M acquisition (announced May 14, 2026). KP-backed multi-cloud networking; Leigh co-published the KP perspective: “AI agents are already generating traffic current networks were not designed for: east-west, cross-cloud, policy-driven.” A live win on her desk — congratulable.
• Avoca full thesis essay (“Bringing AI to the backbone of the real economy,” Apr 27, 2026) + Fortune feature same day. Avoca now $1Bvaluation,$125M raised. The most init.inc-aligned KP text — verbatim hooks: “another part of the economy, far larger and far less digitized… the businesses that keep the physical world running… almost none of it has been touched by software”; “the ROI… is extremely clear and verifiable shortly after first use”; “core infrastructure for a massive segment of the economy that has historically been overlooked by Silicon Valley.”
• Fund context: KP closed $3.5B\ast \ast new(Mar24,2026)=\ast \ast KP22$1B early-stage + $2.5B growth; Leigh is one of the named early-stage partners. Also amplifying new partner Aatish Nayak (ex-Harvey first PM, joined ~May 19).
• 2026 screen (synthesized from her current activity): AI absorbing labor in overlooked real-economy verticals (Avoca), rebuilding entrenched enterprise categories AI-natively (Roadrunner), infra for the AI era (Alkira/Neon/Windsurf) — unifying filter = immediate, verifiable ROI / real day-one demand. Her Dec-2024 “agents don’t work reliably” has resolved into a “where do agents actually work” filter (vertical scope + integration depth + measurable outcome).
• ⚠️ Recency correction: the much-quoted “20k followers / weekly SF poker night / wedding-speech” tweets are 2023–2025, NOT current (snowflake-decoded: Dec 2025, Jun 2025, May 2023). Her genuine last-60-day timeline is mostly portfolio/KP amplification — don’t reference those as “this week.”
• init.inc fit (sharpened): Avoca is a near-one-to-one analog — position init as “Avoca for IT” (AI workforce for the IT/MSP backbone, selling the verifiable outcome “your IT runs”). Conflict check: CLEAN — no AI-native IT-services / ITSM / MSP / helpdesk bet in her book or (surfacing in search) KP’s; closest direct competitors Treeline ($25M Series A, a16z, Mar 31 2026) and SuperOps (March Capital) are not KP-backed → the IT-ops tile is open.

Cross-check vs existing wiki

• ✅ Founders Fund tie: Leigh is an ex-FF Principal — consistent with Founders Fund (our clean-lane firm) and overlaps Ev Randle’s FF-then-KP arc (both passed through FF + KP, different years).
• ⚠️ Avoca: she led Avoca’s Series A — Avoca is market-analog (services-economy AI front-office), not a direct competitor → adjacency/pattern-match, not a clean-lane block. Favorable signal: she already underwrites the init.inc pattern.
• ⚠️ CRM correction: “Citadel & Two Sigma” → Jane Street (see Background). Surface to the principal before any in-person reference.
• ✅ Kushal’s page already records the Kushal→Leigh edge (“Kleiner Perkins / Leigh Marie Braswell — Follow Up Phase”). Consistent.

Related

• Kleiner Perkins · Nadia Cochinwala · Kushal
• Founders Fund (alum) · Ev Randle (parallel FF→KP arc)
• Avoca (led Series A — market-analog)
• cap-table-patterns-across-startup-competitors