Nov 18, 2025 · 1h 14m · neon-show
What 20 Years of Investing taught Somesh Dash of IVP after backing Perplexity, Figma, Dropbox
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this in-depth conversation on The Neon Show, IVP General Partner Somesh Dash joins host Siddharth Ahluwalia to discuss the dynamics of the US-India innovation corridor, evaluate frontier AI investments like Perplexity and Abridge, and outline disciplined venture capital strategies for navigating impending market corrections.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
Siddhartha as informed peer 3.3 Guest teaching 3.8 Guest disagreement 0.8 Siddhartha pushing back 0.7
speaking balance: gold is Siddhartha, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 1:02:54 Rejecting the contradiction between bubbles and winners
Somesh firmly dismisses the host's premise that predicting an AI correction contradicts expecting $100B winners, asserting that market corrections and extreme power-law outcomes routinely co-exist.
Hardest push from Siddhartha ▶ 1:02:42 Confronting contradictory macro viewsSiddharth directly challenges Somesh on an apparent logical inconsistency: forecasting an imminent AI bubble collapse while simultaneously predicting massive continuous market cap growth for top players.
Biggest teaching moment ▶ 7:15 Masterclass on mini-lateral trade frameworksAfter the host asks what the terms mean, Somesh delivers a detailed breakdown of bilateral vs multilateral trade accords and the economic rationale behind agricultural tariffs in India.
Siddhartha holds their own ▶ 57:12 Asserting the India-US corridor labor advantageSiddharth counters the guest's discussion of unsustainable Silicon Valley engineering compensation by bringing his domain expertise on cross-border venture economics, explaining how the India corridor insulates startups from massive US payroll bloat.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Siddhartha as informed peer | Guest teaching | Guest disagreement | Siddhartha pushing back | Why |
|---|---|---|---|---|---|---|
| Silicon Valley's Immigrant Foundation and Indo-US Heritage | 2 | 4 | 1 | 0 | Somesh delivers an extensive historical overview of immigrant contributions to Silicon Valley, including early 20th-century Indian railway workers. Siddharth primarily listens, asking brief clarifying questions about historical demographics. | |
| Geopolitical Dynamics: Mini-Lateralism, Trade, and Bilateral Ties | 2 | 5 | 1 | 0 | The host asks for definitions of multilateralism versus mini-lateralism, allowing the guest to lecture in detail on modern geopolitical trade frameworks and bilateral diplomatic history between the US and India. | |
| The Emergence of Perplexity AI and Arvind Srinivas's Vision | 3 | 3 | 0 | 0 | Siddharth brings up Perplexity AI and founder Arvind Srinivas. Somesh explains his early impressions meeting Arvind at WeWork and highlights the company's product strategy with the Comet browser. | |
| Backing Perplexity: Traction, Diligence, and Global Growth | 3 | 3 | 0 | 0 | Siddharth prompts on the traction and valuation at which IVP led Perplexity's Series B round. Somesh explains IVP's diligence process and benchmarking against earlier hits like Snap and Discord. | |
| Fast-Paced Execution and Venture Collaboration in Silicon Valley | 3 | 4 | 1 | 0 | Somesh dispels the media myth that Silicon Valley venture capitalists act like gladiators fighting each other, explaining that multi-stage investing requires long-term collaboration over 8-10 year horizons. | |
| Figma's Growth Journey and Lessons from Enterprise Winners | 4 | 3 | 0 | 0 | Siddharth queries the entry valuation and conviction behind IVP's investment in Figma. Somesh discusses Dylan Field's vision, Ajay Vashee's relationship with Figma's CFO, and post-IPO holding philosophy. | |
| Market Valuations, Big Tech Resilience, and AI Transformation | 3 | 4 | 1 | 1 | Siddharth asks whether public tech markets are overvalued and whether a company must reach $10B to be deemed successful. Somesh reframes success beyond binary valuations using historical case studies like TiVo and Ask Jeeves. | |
| The DoorDash Miss: Misjudged Unit Economics and Category Expansion | 3 | 4 | 0 | 0 | Somesh reflects openly on IVP's anti-portfolio mistake with DoorDash due to over-indexing on early negative unit economics, detailing IVP's 5-out-of-5 conviction scoring framework. | |
| The Golden Era of Vertical AI and the Shift in Media Consumption | 4 | 3 | 0 | 0 | Siddharth inquires about entry metrics like Abridge's $3M ARR, prompting Somesh to outline why vertical AI in healthcare and education offers massive defensible upside compared to pure horizontal foundation models. | |
| Analyzing Frontier Foundation Models: OpenAI and Anthropic | 3 | 4 | 1 | 0 | Siddharth asks if IVP had opportunities to back OpenAI or Anthropic. Somesh explains non-profit structural constraints and argues mega-cap AI labs operate more like quasi-public utilities requiring state-level CapEx. | |
| The Impending AI Market Correction and Historical Cycles | 3 | 5 | 2 | 1 | Somesh warns of an inevitable AI market correction, comparing current hype to the 1999-2001 dot-com bubble and pointing out historical parallels between early search engines and Gen 1.0 AI startups. | |
| Deploying Capital Across Market Cycles and Valuation Discipline | 5 | 4 | 2 | 3 | Siddharth actively pushes back by noting that the India-US corridor allows startups to bypass Silicon Valley's $100M engineering bidding wars, and challenges Somesh on capital deployment discipline across vintages. | |
| Multi-Stage Investment Strategy and Market Liquidity Dynamics | 5 | 5 | 3 | 5 | Siddharth directly calls out Somesh for holding contradictory positions by predicting a bubble burst while claiming portfolio companies will exceed $100B. Somesh rejects the contradiction, clarifying that power law dispersion allows breakout winners even amidst broad downturns. | |
| Five Decades of IVP: True Partnership and Value-Add Support | 3 | 3 | 0 | 0 | Siddharth asks about the structural durability of IVP over five decades. Somesh details the firm's flat one-partner-one-vote governance model and specialized in-house talent teams. | |
| The Rising Power of the India-US Innovation Corridor | 4 | 3 | 0 | 0 | Siddharth asks if IVP plans to invest directly into domestic Indian consumer companies. Somesh clarifies IVP's focus on backing Indian-founded companies targeting global enterprise scale. |
Statements from this episode (34)
Dash: Recent tensions will spur US-India strategic dialogue
“I'm very bullish, though, that this is going to cause the right strategic dialogue to happen, you know, between India and the United States in the coming months.”
Dash: US defense sector seeks startup alternatives to Russian supply
“The defense sector, which you and I were just talking about, is extremely focused on finding the right primes and startups to actually become a second source to Russia.”
Dash: Perplexity had 20 employees when IVP first met founder
“I believe the company was about 20 employees. They had just launched early versions of the Perplexity product. It was still in beta, and it was still pretty small”
Dash: Perplexity will be preloaded on Airtel phones in India
“I mean, I encourage everybody, actually, I know in India, they've made an announcement on Airtel, like they're going to be preloaded on a lot of the Airtel subscribers phones,”
Dash: IVP led Perplexity's Series B at a $500M valuation
“I believe it was in the fall of 2023, we led the series B round. It was around the price, as you said. I haven't done the exact dilution math, but somewhere around there.”
Dash: Perplexity's early engagement rivaled Discord and Snap
“What I do remember though is we hadn't seen that kind of daily active usage and organic engagement since we had looked at Discord or at Snap.”
Dash: IVP invested in Twitter, Snap, and Discord under $1B valuations
“We were Series B investors in Twitter and Snapchat a little bit later in Discord, but all of which we did well below a billion dollars in venture price.”
Dash: AI startups must rediscover product-market fit every six months
“Every six months you're rediscovering product market fit, whether you're a startup or you're a large company.”
Dash: OpenAI and Anthropic will stop hosting model launch parties by 2028
“Do we think in three years, OpenAI and Anthropic are going to keep doing launch parties for each model? No, at that point, it's just going to be a product release, right? And you're not going to see behind the scenes which model is it five, is it 15, is it 25,…”
Dash: IVP and Elad Gil co-led an Abridge round
“We just co-led a bridge together, you know, about a, about nine months ago at one of the rounds there.”
Dash: IVP typically invests two years after first meeting a founder
“We typically invest about two years after we meet a founder or get to know them.”
Dash: Three IVP portfolio companies entered the S&P 500 in 14 months
“We had three companies in the last 14 months enter the S&P 500. CrowdStrike, Datadog, and Coinbase.”
Dash: Current US public equity markets are not overvalued
“And right now, do you believe the current public markets are overvalued in the US? Somesh Dash: I do not, no.”
Dash: IVP passed on DoorDash in 2017 over negative margins
“The reason we missed it was, and I take responsibility, when we looked at it, I think it was, ah, the summer of 20 17 or 18, At that time, they were launching in cities, and the economics were not great. There were negative gross margins. There were not too ma…”
Dash: DoorDash expanded delivery TAM past early $20B estimates
“In a category that most people, when I looked at it, most people said the TAM for restaurant delivery is twenty billion max, right? They were all wrong. He expanded the category. He expanded the use cases. He changed user behavior.”
Dash: IVP diligences 200 startups annually and invests in 12 to 14
“See for us, I think what we look at about a thousand a year. We seriously diligence and do work on 200, and we only as a firm do 12 to 14, each of us individual partners is one to two, right?”
Dash: IVP co-led CrowdStrike investment at a $3B valuation
“We didn't think when we backed CrowdStrike, you know, our entry price was three billion in the big round that we did. We co-led that round.”
Dash: Ambient AI medical scribing accuracy is approaching 100%
“Now a doctor says, hey, Siddharth, would you mind if I record this conversation? Those conversations now have an accuracy rate that's getting close to a hundred percent.”
Dash: Vertical AI will create $10B ARR companies in 10-15 years
“I'm pretty sure we're going to see five to ten billion dollar ARR companies in healthcare, education, defense, and retail. In the next 10 to 15 years that are not going to be head-on competing with the same companies that are competing code generation.”
Dash: IVP invested in Abridge at $3M ARR in January 2024
“A bridge was January, 20, 24. It's funny how much has happened. I still remember certain things. I believe when we invested the company was only at three million of ARR.”
Dash: IVP avoided early OpenAI due to its capped-profit structure
“The original structure of OpenAI was such that I think it would have been difficult for us given the cap on the profits as well as the nonprofit structure.”
Dash: Perplexity is IVP's core bet in consumer AI
“We are investors in perplexity and obviously like for that's been our bet in that space and consumer.”
Dash: OpenAI and Anthropic could go public tomorrow based on financials
“They're almost quasi public companies. They could go public tomorrow based on the financials that we're seeing.”
Dash: AI is moving too fast and a market correction is near
“What is inevitable though, if you look at that six year period between the Netscape IPO and the downturn, and you say everything's happening at twice the speed, we're probably pretty close to being due for an AI correction.”
Dash: Most Gen 1.0 AI startups will not reach $10B valuations
“I am pretty convinced that a lot of the gen one dot O companies won't get there. They won't become the ten billion dollar plus outcomes.”
Dash: 2025 VC cohort will see overinflated entry valuations like 2021
“The 2021 cohort, The entry prices for every venture campus, almost by definition, was overinflated, right? I think the twenty-twenty-five cohort's gonna have some of that too, because you're seeing a lot of the early markups ratchet up people's enthusiasm, the…”
Dash: IVP deploys roughly $500 million of capital annually
“We deploy about call it half a billion a year.”
Dash: Hello Heart is approaching $100M ARR after 2021 investment
“When we invested, they were at ten million of ARR. They're on the cusp of breaking through a hundred”
Dash: M&A and PE will drive startup liquidity over IPOs
“I think what we're going to see more of is acquisitions by private equity firms and other startups, mergers, and of course, strategics being a form of liquidity versus just the IPO.”
Dash: IVP strictly deploys funds over three years despite market hype
“Every Monday they said, I know there's a temptation to want to put more money in, but we told our LPs we have to invest over a three year time horizon. We're on that time horizon. We don't want to go faster. So we've been pretty judicious with how we do that.”
Dash: IVP operates as an equal-vote partnership with no managing partner
“There is no CEO of IVP. There's no kind of managing partner. There's no super management company. It's a very simple structure, which is in every fund, the partners, one partner, one vote, it's what a true partnership means, right? Even if you're young and sta…”
Dash: Jyoti Bansal is the Steve Jobs of infrastructure software
“Probably in some ways almost like the Andy Bechtelschein or Steve Jobs equivalent in, like, infrastructure software and what he's done with three companies in the last decade.”
Dash: Over half of headcount at some IVP portfolio companies is in India
“And if you look at the headcount of some of our companies, more than half are actually in India, right? And it's not back office. It's a lot of the innovation, the founders, they live in India.”
Dash: Proliferation of venture capital firms will not degrade overall returns
“I know a lot of people are like, there's so many venture firms. This is horrible. Returns are going to go away. I'm actually, I think it's the opposite. I think the markets are efficient enough where everybody's got a specialization and a niche, and you get th…”