Last month, in Japan, venture capitalist Deedy Das was supposed to be off the grid, focused solely on proposing to his girlfriend. Two nights into the trip, the deal flow found him anyway.

A founder he’d been getting to know over eight months texted to say he was kicking off a raise. Other investors were circling, but the founder wanted Das and his firm, Menlo Ventures, at the table. So between temple visits and kaiseki meals — and once when he told his girlfriend he was soaking at the onsen — Das coordinated with colleagues back home.

The day after he returned to California, the founder pitched Menlo’s partners. That evening, Das — newly engaged — and the founder celebrated a handshake agreement over Japanese food at Ozumo.

In only a year and a half working in venture, Das has quickly become one to watch. He joined Menlo after stints as an engineer at Facebook and Google, and as an early employee at the enterprise search company Glean, where he helped scale the business from having no product to a $2.2 billion valuation. In June, that figure jumped to $7.2 billion.

At Menlo, he’s carved out a role at the cutting edge of artificial intelligence, infrastructure, and enterprise software investing. He helped launch the firm’s $100 million Anthology Fund, a joint venture with Anthropic, and earned a reputation for his unusually technical diligence. Now, with his promotion to partner, Das is part of a new generation climbing venture capital’s ranks.

Increasingly, firms are looking to leaders who come from the trenches of engineering and startups, not just the spreadsheet jockeys groomed in banking. In today’s market, where breakthroughs come wrapped in jargon and white papers, technical fluency isn’t a nice-to-have. It’s practically the job description.

Menlo is bulking up its roster with investors who bring technical bona fides. This year, it hired cybersecurity guru Matt Kraning, who sold his startup to Palo Alto Networks in a deal topping $1 billion.

Tim Tully, a general partner and Splunk’s former chief tech officer, previously told Business Insider, founders “want investors who they can talk shop with.”

Inside track

Deep learning and natural language processing have always been central to Glean’s products. But as models advanced gradually, then all at once, CEO Arvind Jain assembled a dedicated “tiger team” of engineers to fold the latest capabilities into the platform.

Das led that effort, steering the company beyond its original Google-like search interface into something that digests data, interprets it, and can perform tasks on a user’s behalf. Today, the company is on pace to hit $250 million in annualized revenue by the end of the year, up from $100 million in 2024.

Jain says the systems and products that Das, who left Glean in 2024, helped create are “still central to how our customers use Glean.”

“I’ve known Deedy for many years,” Jain said, “his impact at Glean was significant as a founding engineer — from shaping our products and raising the bar on engineering excellence to mentoring teammates and helping build the culture that carried us through scale.”

Das says his background gave him an early read on OpenRouter, which offers developers a single platform to access multiple large language models. Having built a version of the same tool at Glean, he knew the demand would be real.

When he met founder Alex Atallah, Das didn’t hedge: “This is absolutely going to be useful. Model velocity is not going to stop. Whenever you’re ready to raise, we’re in.” He laughs now, knowing that he didn’t have the authority to promise that deal, but was willing to bluff. Menlo first backed the startup with a modest check through the Anthology Fund before doubling down to lead the $40 million Series A in June.

Over the past year, Das says OpenRouter has scaled from processing about 10 trillion tokens annually to more than 250 trillion — a 25x jump in throughput. Because its business model ties revenue directly to usage, that surge suggests revenue has grown in lockstep.

Das’s other investments include Wispr Flow, an app that turns messy speech into polished writing, and Goodfire, a research lab focused on understanding why machine learning models behave the way they do. Both came in through the Anthology Fund before Menlo followed on with more capital. These deals and others earned Das a spot on Business Insider’s 2024 list of the rising stars of the venture capital industry.

“Deedy is someone who has conviction in bold ideas where others may be skeptical,” says Goodfire CEO Eric Ho. “He has an ability to see the future in a way.”

The same technical rigor that helps Das land deals also cuts through hype.

A few months ago, Das looked at a pitch from a buzzy startup raising a round and thought its cohort retention curve, a measure of product stickiness, looked too good. He was right. After pulling an Excel file from the company’s data room, he ran the numbers through Claude Code to re-graph the curve. He realized the startup had juiced the math to make retention look stronger than it was.

Between sourcing deals and reading white papers, Das also leans on AI to boost his own productivity. He uses Wispr to send texts and clear his inbox faster. With Claude, he vibe-coded a news aggregator that pulls updates from his preferred sources. He’s built one agent to research people on his calendar, and another that scrapes LinkedIn to find out who’s starting companies.

“It’s kind of strange,” Das laughs. “We invest in the cutting edge, but most venture firms don’t use any cutting-edge technology to do their actual work.”

Have a tip? Contact this reporter via email at mrussell@businessinsider.com or Signal at @MeliaRussell.01. Use a personal email address and a non-work device; here’s our guide to sharing information securely.

Last month, in Japan, venture capitalist Deedy Das was supposed to be off the grid, focused solely on proposing to his girlfriend. Two nights into the trip, the deal flow found him anyway.

A founder he’d been getting to know over eight months texted to say he was kicking off a raise. Other investors were circling, but the founder wanted Das and his firm, Menlo Ventures, at the table. So between temple visits and kaiseki meals — and once when he told his girlfriend he was soaking at the onsen — Das coordinated with colleagues back home.

The day after he returned to California, the founder pitched Menlo’s partners. That evening, Das — newly engaged — and the founder celebrated a handshake agreement over Japanese food at Ozumo.

In only a year and a half working in venture, Das has quickly become one to watch. He joined Menlo after stints as an engineer at Facebook and Google, and as an early employee at the enterprise search company Glean, where he helped scale the business from having no product to a $2.2 billion valuation. In June, that figure jumped to $7.2 billion.

At Menlo, he’s carved out a role at the cutting edge of artificial intelligence, infrastructure, and enterprise software investing. He helped launch the firm’s $100 million Anthology Fund, a joint venture with Anthropic, and earned a reputation for his unusually technical diligence. Now, with his promotion to partner, Das is part of a new generation climbing venture capital’s ranks.

Increasingly, firms are looking to leaders who come from the trenches of engineering and startups, not just the spreadsheet jockeys groomed in banking. In today’s market, where breakthroughs come wrapped in jargon and white papers, technical fluency isn’t a nice-to-have. It’s practically the job description.

Menlo is bulking up its roster with investors who bring technical bona fides. This year, it hired cybersecurity guru Matt Kraning, who sold his startup to Palo Alto Networks in a deal topping $1 billion.

Tim Tully, a general partner and Splunk’s former chief tech officer, previously told Business Insider, founders “want investors who they can talk shop with.”

Inside track

Deep learning and natural language processing have always been central to Glean’s products. But as models advanced gradually, then all at once, CEO Arvind Jain assembled a dedicated “tiger team” of engineers to fold the latest capabilities into the platform.

Das led that effort, steering the company beyond its original Google-like search interface into something that digests data, interprets it, and can perform tasks on a user’s behalf. Today, the company is on pace to hit $250 million in annualized revenue by the end of the year, up from $100 million in 2024.

Jain says the systems and products that Das, who left Glean in 2024, helped create are “still central to how our customers use Glean.”

“I’ve known Deedy for many years,” Jain said, “his impact at Glean was significant as a founding engineer — from shaping our products and raising the bar on engineering excellence to mentoring teammates and helping build the culture that carried us through scale.”

Das says his background gave him an early read on OpenRouter, which offers developers a single platform to access multiple large language models. Having built a version of the same tool at Glean, he knew the demand would be real.

When he met founder Alex Atallah, Das didn’t hedge: “This is absolutely going to be useful. Model velocity is not going to stop. Whenever you’re ready to raise, we’re in.” He laughs now, knowing that he didn’t have the authority to promise that deal, but was willing to bluff. Menlo first backed the startup with a modest check through the Anthology Fund before doubling down to lead the $40 million Series A in June.

Over the past year, Das says OpenRouter has scaled from processing about 10 trillion tokens annually to more than 250 trillion — a 25x jump in throughput. Because its business model ties revenue directly to usage, that surge suggests revenue has grown in lockstep.

Das’s other investments include Wispr Flow, an app that turns messy speech into polished writing, and Goodfire, a research lab focused on understanding why machine learning models behave the way they do. Both came in through the Anthology Fund before Menlo followed on with more capital. These deals and others earned Das a spot on Business Insider’s 2024 list of the rising stars of the venture capital industry.

“Deedy is someone who has conviction in bold ideas where others may be skeptical,” says Goodfire CEO Eric Ho. “He has an ability to see the future in a way.”

The same technical rigor that helps Das land deals also cuts through hype.

A few months ago, Das looked at a pitch from a buzzy startup raising a round and thought its cohort retention curve, a measure of product stickiness, looked too good. He was right. After pulling an Excel file from the company’s data room, he ran the numbers through Claude Code to re-graph the curve. He realized the startup had juiced the math to make retention look stronger than it was.

Between sourcing deals and reading white papers, Das also leans on AI to boost his own productivity. He uses Wispr to send texts and clear his inbox faster. With Claude, he vibe-coded a news aggregator that pulls updates from his preferred sources. He’s built one agent to research people on his calendar, and another that scrapes LinkedIn to find out who’s starting companies.

“It’s kind of strange,” Das laughs. “We invest in the cutting edge, but most venture firms don’t use any cutting-edge technology to do their actual work.”

Have a tip? Contact this reporter via email at mrussell@businessinsider.com or Signal at @MeliaRussell.01. Use a personal email address and a non-work device; here’s our guide to sharing information securely.

Consumer tech investing has fallen out of fashion in the last four years. A small but vocal cohort of VCs is rallying to bring it back.

The VC boom and subsequent bust of the early 2020s hit consumer startups particularly hard. Many investors, burned by the crash, set their focus on startups serving businesses that could deliver reliable contract-based revenue — and, in the past two years, that could more readily capitalize on venture’s AI gold rush.

But other VCs believe AI will change the game for startups selling to consumers, too.

“It’s always darkest before the light and the pendulum for consumer is now swinging back, due to AI,” said Nicole Quinn, a consumer-focused VC at Lightspeed Venture Partners, in an X post in January.

The especially bullish investors are doubling down on consumer to “buy the dip” as other VCs stay away. Mercedes Bent stepped down from her role at Lightspeed Venture Partners earlier this year to raise money for a firm she’s launching alongside former New Enterprise Associates partner Vanessa Larco, called Premise Venture Partners, which will invest exclusively in consumer startups.

“Everyone’s moving away from consumer investing at the wrong time. It’s classic loss aversion,” she told Business Insider. In Bent’s view, their loss is her gain.

In addition to Premise, consumer-focused funds like Hobart Ventures (founded by former Raya and Dispo exec TJ Taylor) and Parable (founded by ex-A16z partner Anne Lee Skates) have launched in the past year. Other VC mainstays, like Menlo Ventures, Maveron Ventures, or Bessemer Venture Partners, are continuing to place bets in the category.

There’s no shortage of hurdles in their path.

According to Silicon Valley Bank, the amount closed by VC funds listing consumer as a core focus hit a seven-year low in 2024, totalling about $9 billion, compared to $65 billion in 2021. Meanwhile, consumer startups on Carta raised 47% less cash in the first quarter of 2025, according to Carta, which said in a June report that the sector appears to be stabilizing at a “new normal” for funding that’s significantly down from 2019 to 2022.

Consumer startups also risk flaming out after a moment in the spotlight. Look no further than 2021 breakout social network Clubhouse, which raised over $100 million from VCs and received a $4 billion valuation, only to fizzle in popularity after companies like Twitter (now X) built copycats. Some consumer startups, like TikTok rival Flip, end up shutting down.

But investors, including at mega-firms like Andreessen Horowitz, increasingly think AI could usher in new consumer business models that enable unprecedented growth.

“This shift allows consumer companies to scale in ways that were once impossible — and to become big businesses (on a revenue basis) in months, not decades,” said Olivia Moore, an A16z partner focused on AI investments, in a September blog post.

Consumer’s big money question

Some remain skeptical about how consumer tech startups can make money as Big Tech giants dominate.

Brian Sugar, the cofounder of media brand PopSugar and founding partner of Sugar Capital, doesn’t think the industry has figured out how to solve the monetization and distribution problems created by the last consumer tech boom.

“If you have a consumer hit like ChatGPT, which is truly a new thing, you can get that level of traction,” Sugar told Business Insider. “But I don’t think AI all of a sudden changes what’s happened in the consumer technology landscape.”

Blockbuster AI tools have captured the zeitgeist, from OpenAI’s ChatGPT to Anthropic’s Claude, but many of these companies appear to rely more heavily on revenue from other businesses rather than subscriptions from individual users, with only about 3% of consumers paying to use premium AI services, according to Menlo Ventures.

But investors buying into consumer tech are betting that LLMs will bring the same kind of shift to user behavior, and thus to monetization opportunities, that the iPhone brought in the late 2000s.

If search traffic moves from Google to ChatGPT, consumer tech companies may no longer need to optimize for Google’s algorithms, but for OpenAI’s.

“Today, I’m telling all of my companies, figure out distribution via ChatGPT, Claude, Perplexity. They’re not optimized yet — those companies haven’t even begun to take a profit from the distribution they’re creating. But one day they will, sometime in the next five years,” Bent said. “If you can’t be the entity that controls the distribution, then optimize your customer acquisition costs by figuring out what’s the latest up-and-coming platform.”

Subscription models have become familiar to users, too. Half of Americans online have at least four paid monthly subscriptions, according to Forrester. Consumer AI startups are catching on and charging monthly for their products to secure more predictable revenue. Some are even implementing usage-based models, with higher costs or the option to buy additional “credits” for more active users.

For instance, Rosebud, an AI journaling startup backed by Bessemer Venture Partners, is free to use but charges users a $13 a month subscription fee for its more in-depth AI tools, like long-term memory or voice transcription. Image and video generator Krea, backed by A16z, offers plans from $10 to $60 a month based on the number of tasks users want the model to perform, and the option to buy additional credits exceeding those tasks.

Natalie Dillon, a partner at consumer-only VC firm Maveron, said she expects more consumer tech companies to diversify their businesses to take advantage of new advertising options that’ll appear if and when OpenAI and its LLM competitors begin running ads for free users.

“Right now, a lot of these businesses are very subscription-based because that’s a model that is comfortable with consumers today. I think over time we’ll actually see more ad-driven startups, but that market and that network just haven’t been built yet,” she said.

Investors are oggling white spaces in consumer AI

Capturing consumers’ attention is harder than ever, but startups may yet be able to attract fresh eyes — and checks — in certain segments. Consumer startups peddling AI assistants, for example, raised a total of $1.3 billion from US-based VCs in 2024, according to Silicon Valley Bank.

Granola, an AI assistant that takes notes in meetings, raised $43 million in May. Alta, an AI assistant that helps consumers choose what to wear, announced an $11 million seed fundraise in June.

Amy Wu Martin, a partner at Menlo Ventures (which led Alta’s seed round), said she anticipates that a “rich ecosystem of specialized consumer AI products” across categories like healthcare, shopping, and entertainment will emerge in the “next couple of years.”

In gaming, startups like Northzone-backed Hidden Door, launched in August, are making fiction interactive with AI. Whoop and Oura, both valued at billions of dollars in their latest fundraises, use AI to track and analyze health and wellness data collected via their wearables. Social networks and dating apps are also leveraging AI and securing early-stage investments or spots in buzzy accelerators like A16z Speedrun, such as dating startup Sitch.

“Every time we’ve had a major enabling technology show up, it’s led to the rise of a new social network dynamic,” said Kent Bennett, a partner at Bessemer Venture Partners.

Not every consumer startup needs AI, however. Sugar said the best-performing consumer company in his portfolio is Locket, a social media app that lets users share photos with their friends and family via a widget on their phones’ home screens. The app, which doesn’t rely on AI, went viral on TikTok in early 2022 and has become especially popular with preteen users, he said.

Each generation tends to stick with the consumer tech platforms they grew up with — Facebook for Gen X, Instagram for millennials, and TikTok for Gen Z, Sugar said. He thinks Gen Alpha, the next generation of consumers, could be up for grabs for the next startup that can capture and retain their attention.

That proposition, while intuitive, comes with plenty of risk. Character AI shows just how fraught such a contest can be. The company, which notched a $2.7 billion licensing-and-acquihire deal with Google last year, has drawn millions of young users to its AI-generated companions. But it’s also come under fire for inadequate safeguards, including a lawsuit that claims one of its chatbots played a role in a teenager’s suicide.

Just this month, the FTC launched an inquiry into companies like Character AI and the impact AI companions are having on children.

“The user-created characters on our site are intended for entertainment and we have prominent disclaimers in every chat to remind users that a character is not a real person and that everything a character says should be treated as fiction,” a spokesperson for Character AI told BI in a statement. They added that the startup has “put a tremendous amount of resources” toward trust and safety.

So, is consumer back?

Dillon said that consumer investing boomerangs in and out of vogue: “We’re in a pattern that we’ve seen before.”

Interest in consumer typically retreats as a result of macroeconomic pressures, she said. When a new technological shift arrives — right now, that’s AI —excitement returns.

Meanwhile, Bennett said he thinks some investors have overcorrected in ditching consumer tech altogether.

“There are enough people who were burned in the last 10 years investing in direct-to-consumer stuff that may have told themselves that consumer stinks,” he said. “There’s a little bit of that hangover from that top-down lesson.”

But when startups do get it right, the potential upside is enormous.

“It’s really hard, but when you hit it, you instantly have access to trillion-dollar markets,” he said. “If you want to win the consumer over as a product, you just have to show up with something that is a miracle.”