← Back to Startups List
Reflection AI logo

Reflection AI

reflection.aiadded 2026-09-02
Fundraising pace far outruns product pace: three rounds (seed/A, B, C) pushed valuation from $545M to $27.5B in about a year, while the flagship model doesn't ship until early 2026 — the market is pricing the team's pedigree and narrative, not proven revenue

Worth-Joining Index · Scorecard

v2.1 ALPHA · 2026-08

Career Score

/100

Coverage

51%

Verified

0%

Stage cohort

Early

Vibe Check

🎪 community voting coming soon

Momentum
4/4
Efficiency
unknown
Backing
7.6/8
Market
7.7/11
Product
5.5/11
Upside
4/8

Latest valuation

$8B (Series, 2025-10)

Series B · 2025-10

Total raised

~$2.13B total raised (2025-10)

as of 2025-10

Team size

~60 people

figures vary
AI编程开源大模型智能体Founded · 2024Stage · Growth (Series C)HQ · New YorkSize · ~60 people (2025-10)

Traction · Key numbers

Valuation paceValuation ~50x in 12 months: $545M → $27.5B Publicly verifiableas of 2026-03
Headcount~60 people (Oct 2025)~ 3rd-party estimateas of 2025-10
Total raised~$2.13B raised as of Oct 2025, plus a new $2.5B round in Mar 2026 Publicly verifiableas of 2026-03

Funding

Total raised

~$2.13B total raised (2025-10)

Rounds

2

Valuation pace

$545M

B

$2B

Series B · 2025-10 · valuation $8B post (Series, 2025-10) · led by Nvidia, Sequoia Capital (with Eric Schmidt, Citi, 1789 Capital, Lightspeed Venture Partners, DST, GIC, CRV, B Capital participating)

With: Sequoia Capital, Citi, GIC, 1789 Capital, B Capital, DST, CRV, Disruptive · 15x valuation increase from Series A; individual investors Eric Schmidt and Eric Yuan also participated

S

$130M

Seed + Series A · 2025-03 · valuation $545M

Combined $25M seed and $105M Series A as the company emerged from stealth

Fundraising pace far outruns product pace: three rounds (seed/A, B, C) pushed valuation from $545M to $27.5B in about a year, while the flagship model doesn't ship until early 2026 — the market is pricing the team's pedigree and narrative, not proven revenue

Nvidia-led with strategic co-investors (Eric Schmidt, Citi, 1789 Capital) suggests this is compute/ecosystem positioning as much as financial betting — heavy infra dependency ahead

Sequoia's repeat participation across rounds signals strong conviction in the founding team (DeepMind's Gemini reward-modeling lead, AlphaGo co-creator) — a classic 'bet on the people' pattern

Founders

Misha Laskin CEO

Led reward modeling for DeepMind's Gemini project

Ioannis Antonoglou 联合创始人

Co-created AlphaGo, which defeated the world Go champion in 2016

Reality · Working here

H1B / PERMUnknown
Remote policyUnknown
Comp modelUnknown
GlassdoorUnknown — work here? Help us fill this in →
Equity liquidityUnknown
Hiring statusUnknown

Signals & Risks

CompetitorsAnthropicOpenAIDeepSeekMistral AI
ValuationValuation jumped from $545M to $8B in 7 months while the flagship product has yet to launch at scale, risking overextended expectations
Product deliveryThe first flagship frontier model won't ship until early 2026, so commercial validation lags well behind the fundraising pace
CompetitionThe open-frontier-model space faces squeeze from both low-cost competitors like DeepSeek and top labs like Anthropic and OpenAI

Editorial ✎

Good fit for

· Top RL/pretraining researchers — the team (DeepMind's Gemini reward-modeling lead, AlphaGo co-creator) is rare in spanning both frontier models and agent products

· Believers in the 'open frontier model' thesis — if you think the ecosystem needs an open frontier lab beyond Meta, this is currently the best-capitalized bet

· People comfortable with high uncertainty and pure optionality — valuation has outrun product shipping by a wide margin, so joining early is a venture bet, not a stable offer

Not for

· Anyone who needs to see a mature product and customer case studies before deciding — Asimov and the flagship model have no large-scale commercial validation yet

· Anyone sensitive to compute/infrastructure supply-chain risk — Nvidia is both investor and key supplier, creating complex incentive entanglement

Ask in the interview

1. How real is flagship-model slip risk? Has the early-2026 release timeline already been pushed internally?

2. At a $27.5B valuation, what's the actual ARR/paying-customer count, and does it match the pace of valuation growth?

3. How does the open-source strategy compete against low-cost players like DeepSeek and closed frontier labs like Anthropic/OpenAI — what specifically is the moat?

Data changelog

Community corrections → editorial review → applied, fully on the record

2026-09-02
团队规模 filled in as ~60 people (2025-10)

data as of 2025-10 · Source: TechCrunch

Publicly verifiable
2026-09-02
最近一轮领投机构 filled in as Nvidia、Sequoia Capital (with Eric Schmidt, Citi, 1789 Capital, Lightspeed Venture Partners, DST, GIC, CRV, B Capital participating)

data as of 2025-10 · Source: TechCrunch

Publicly verifiable
2026-09-02
累计融资额 filled in as ~$2.13B total raised (2025-10)

data as of 2025-10 · Source: TechCrunch (derived: $130M Mar 2025 seed round + $2B Oct 2025 round)

~ 3rd-party estimate
2026-09-02
最近一轮估值 filled in as $8B post (Series, 2025-10)

data as of 2025-10 · Source: TechCrunch

Publicly verifiable
Spot an error? Submit a correction → — you can correct facts, never negotiate scores.

Want to talk about this company?

Community members may work here or have interviewed — corrections and additions welcome too

Join the discussion