Emma Qian

I've been building AI companies since 2014. Node — an AI company using LLMs years before ChatGPT existed (acquired by SugarCRM). Fireside — a creator operating system co-founded with Mark Cuban. Now I'm General Partner at Digital Indies, and this week my AI agents built 5 autonomous systems while I was in meetings: → A company from scratch. No instructions, no check-ins. → A self-learning engine that monitors markets and teaches itself. → An outbound pipeline. Finds leads, writes emails, sends, follows up. 0% bounce. → A trading portfolio with autonomous signal detection. → A content system. One idea → native posts for 5 platforms. For the first time, it's faster to build than to talk about it. That's what we're building at Digital Indies — an operating system on top of AI agents that takes the technical guesswork out. So founders, creators, and operators can get value without worrying about the rest. The future isn't learning to code. It's learning to architect. What would you build if your AI agents could work while you sleep?

42

2 Comments

$100 M for one engineer sounds outrageous—until you run the numbers for Meta ...or look at the GPU invoices Back of the Napkin Math and sources in comments. Revenue math • Meta booked $164.5 B in ad revenue during 2024 on ~4 B monthly active people—about $41 ARPU.¹ • A $100 M package amortises to $25 M per year. • Raising ARPU by just 0.02 % (+ $0.008) adds ≈ $32 M/yr—already positive ROI. Compute math • Management now guides $64–72 B of 2025 capex—nearly all pointed at AI infrastructure.⁴ • Databricks shows Llama 3.3 serving optimisations can slash inference costs by up to 80 %.³ • Kernel-level tricks like FlashAttention-2 double training throughput.⁵ One engineer who lands breakthroughs like these can redirect hundreds of millions straight to the bottom line. Tiny teams, massive upside • < 200 people turned Reels into a $10 B run-rate business in 18 months.² • A handful of PhDs once tweaked News-Feed ranking and grew session length 1 %—worth billions in extra ad impressions. Talent arbitrage, not extravagance •Boards routinely approve $19 B acquisitions (WhatsApp) or $30 B buybacks on shakier forecasts. A nine-figure “micro-M&A” for a human asset who compounds value every sprint is cheap insurance at Meta’s ≈ $1.6 T market cap. •And yes—Zuck is really dangling nine-figure offers to top AI talent.⁶ Bottom line In today’s AI land-grab, the real risk isn’t overpaying for genius—it’s under-recruiting and watching competitors train on your users. Agree? Think $100 M is still too rich? Convince me below. #AIeconomics #TalentCapital #Meta

58

8 Comments

Zynga sold for $12.7 billion. Its founder says startups fail due to this: They try to innovate too early (unless its deep tech). Mark Pincus calls it the “All new fails” rule: 1/ Don’t start with “new”: Start with what’s already proven to work. Copy it legally, study it deeply. 2/ Proven means 10 out of 10 users say: “Yeah, that works.” If they don’t, it’s not proven. 3/ Founders fail because they skip the proven phase: They chase novelty before they’ve earned the right to innovate. 4/ The Zynga story began with poker: Same table, same felt, same gameplay, nothing original. 5/ Then they made it better: No downloads. Just click and play. Friction cut in half. 6/ “Better” means 10 out of 10 users agree: Not your team, not your investors, real users. 7/ Only after proven and better do you earn “new”: That spark that surprises and delights users emotionally. 8/ Zynga’s “new” was showing your friends’ faces: It made games social, not solitary. 9/ Forget MVP, build a minimum viable idea: A concept that hits emotional resonance before metrics. 10/ Seek true signal: You’ll feel it viscerally before you can measure it. 11/ Silicon Valley worships originality: But originality without resonance is noise. 12/ Better founders are great students first: They copy what works before they try to change it. 13/ Most products die from founder ego: They’re “new” before they’re useful. 14/ Users don’t care about novelty: They care about ease, trust, and delight. 15/ “Proven → Better → New just works. And it built a $12.7B company. 📩 Get ai + vc intel in your inbox via my newsletter Startup Riders, link under my name ☝️ 📌 Source: Mark Pincus on Why Most Startups Fail a16z #ai #llms #agent #startups #founder #vc

40

4 Comments

Most founders think the hardest part of deep tech is invention. It’s not. The real bottleneck is distribution. A breakthrough without distribution? A dusty research paper. A breakthrough without reshaping? Just noise. I was enlisted on a content distribution + reshape problem. The challenge: take insane complexity → reshape it → distribute it → move adoption. That’s not “marketing.” That’s infrastructure. Here’s the system I use: 1. Engagement Layer (attention + social proof) Not about leads. About credibility. Reshape deep tech into broad, shareable insights. Examples: “The biggest myth about deep tech startups no one talks about” “Why most AI companies aren’t actually building intelligence” “The meeting that changed how I think about autonomy forever” When strangers see 500+ likes, they assume you’re worth listening to. That’s social proof. Without it, your smartest insight is invisible. 2. Qualification Layer (filter + educate) Now separate casual scrollers from serious prospects. Examples: “Why deep tech founders struggle to raise capital compared to SaaS” “The hidden cost of building AI without simulation” “What investors really look for in deep tech before writing a check” This content doesn’t go viral. It’s a filter. The right prospects engage, comment, or DM. That’s your signal. 3. Conversion Layer (proof + action) Show outcomes. Make the ask clear. Examples: “How one founder cut 18 months off product development using synthetic data” “The exact framework we use to scale deep tech infrastructure into viable companies” “Why we only work with 10 ventures per quarter” Here, metrics aren’t likes. They’re applications, DMs, and deals. Most founders screw this up: Only engagement (vanity, no pipeline). Conversion too early (no trust). Mixing everything (confusing as hell). But the truth is simple: Engagement earns the right to be heard. Qualification earns the right to be considered. Conversion earns the right to be backed. This is the deep tech pipeline: Noise → Signal → Adoption. Stack the layers. Don’t treat content as one-size-fits-all. Treat it as infrastructure. Because when you layer it right, engagement stops being vanity. It becomes enrollment. So tell me—where do you see most founders getting stuck: engagement, qualification, or conversion?

2

A lot of VCs use founder age as a proxy for how well they’ll leverage LLMs and the latest AI. In my experience, the stronger signal is how technical they are. Technical founders understand how LLMs work and can exploit them fully; non-technical founders see magic, not mechanisms.

49

11 Comments

How AI is changing venture building + numbers & best practices of PSL startup studio ($50M AUM studio funds + $180M AUM venture funds): 1) 3 dedicated studio funds: $12M (2015) → $16M (2018) → $20M (2023) for a 25-person build team (salary, sprints, infrastructure). Venture arm: $80M (2018) + $100M (2021) — invests in external startups and PSL studio spin-outs (but never prices later rounds). Geographically focused on the Pacific Northwest. 2) 5-stage company creation funnel: – Ideation (problem hunt, AI research),~4,000 ideas – Validation (customer calls, landing page), 470 validated ideas – Creation (MVP, sales), 50 created products – Spin-out (legal, PSL seed round, CEO equity), 41 spin-outs – Scale-up (external VC funding, recruiting). 37+ scale-ups with external funding. 3) Throughput: 4-6 new spin-outs per year target. Typical time of a startup in the studio: 3–4 months. Fastest build ever: 6 weeks. At any moment: ~2 projects in Creation, 3-5 in Validation. 4) How AI is shrinking Ideation & Validation phases with 2 PSL internally created tools: – Spark – researches the market, auto-drafts decks & landing pages, creates an initial ICP list, then A/B-emails prospects until it gets 10 yes/no signals. Founders start with data, not vibes. – Mesh – maps 70,000+ operator/investor contacts of the PSL team, scores fit, then sends emails and tracks reply rates; the warmest path shows up as a Slack thread the same day. Together they compress “customer proof + talent hunt” to < 24 h instead of 2-3 weeks earlier. 5) Seed checks. Non-AI spin-out: $300-500K from PSL Ventures. AI spin-out: $1.5M co-led ($750K PSL Ventures + $750K AI Studio Fund with Mayfield). Follow-on rounds are syndicated — outside VCs set price; PSL follows its pro-rata. 6) Typical equity split is 50/50 between studio & founding CEO. But can vary a bit based on circumstance (founder track record, founder-idea fit narrative, existing work done by the founder prior to the Studio engagement...) 7) The 4 founder archetypes PSL is partnering with: – Successful repeat founder (typically, 40-50 y.o.) They value the speed & the focus, which is possible due to the studio team + higher conviction in an idea you will bet 5+ years of your life on. – Product Manager to CEO (~25-40 y.o.) They value studio expertise and support as a first-time CEO. They can also find some strong ideas inside the studio if they don't have one. – Young Tech Builder (YTB) < ~30 y. — typically miss and undervalue a new company's business development/sales roles. Studio helps with it. – Corporate exec/domain expert. They know how to get the maximum from the corporate partner but need support as a first-time CEO. Watch the full interview with T. A. McCann – Managing Director @ Pioneer Square Labs to learn insights about venture building with AI and managing a startup studio: youtu.be/YP-i5mhXNiY Bonus from PSL: Would you like to see a demo video of how Spark works? I can DM the link if interested

105

10 Comments

Building a DevTool? These 40+ VCs actively invest in the space-- who are we missing? If you are building developer infrastructure, tooling, platforms, or workflows, these funds have deep technical DNA and consistent interest in dev-first products. A strong place to start. 🚀 Foundational Infrastructure Investors Andreessen Horowitz 1752vc (Formerly Pegasus) CRV Accel Menlo Ventures SignalFire Point Nine SIERRA Ventures 🧠 Early-Stage, Pre-Seed & Founder-Led Funds Afore Capital Precursor Ventures Hustle Fund Founder Collective Bling Capital Great Oaks Venture Capital Flybridge 🧱 Infrastructure, Data & Systems-Focused Initialized Capital Tribe Capital Khosla Ventures Insight Partners General Catalyst Energy Impact Partners 🌍 Platform, Network & Global Funds BoxGroup Global Founders Capital FJ Labs Crosslink Capital Merus Capital Presence Capital 🎮 Vertical & Thematic Specialists (Dev, Security, Gaming, Tooling) 8-Bit Capital BITKRAFT Ventures Konvoy Ballistic Ventures Elementum Ventures ⚡ Accelerators & Hybrid Platforms Techstars Eniac Ventures 645 Ventures Recursive Ventures Uncork Capital Streamlined Devtools win when distribution, community, and product-led growth align. These firms understand that motion. Bookmark this if you are building for developers 👇 #venturecapital #VC

46

8 Comments

On "multiplayer AI": Most people still think about AI as single player software. You prompt a model. It returns an answer. Session over. The emerging opportunity is in multiplayer AI. Systems where many humans and AI agents operate inside the same environment. When that happens, the ** network effects ** look very different from traditional "web2" social networks or "dotcom" marketplaces. They are not simply “more users equals more data.” Instead, the compounding loop looks more like this: 1. More participants create more intents expressed in the system. 2. More intents generate richer interaction traces between humans and agents. 3. Those interaction traces create stronger reinforcement learning signals. 4. Better reinforcement learning produces more capable agents coordinating tasks and decisions. 5. More activity builds deeper shared memory and context across the network. Over time the system accumulates a living operational memory of how work actually gets done. The result is a compounding advantage. The agents understand the environment better. Humans receive better assistance. Workflows become harder to replicate elsewhere. The moat is not just the model. It is the network memory of how thousands or millions of intents get resolved. That is what makes multiplayer AI environments potentially powerful platforms.

8

The most salient narrative in tech today is the end of software. Public software companies have been pummeled, down 20%+ in the last 3 months. But I think the bear case rests on weak logic. There are 3 main categories of concern driving this narrative: 1. Vibe coding means businesses build their own software 2. AI automates away jobs, and seat licenses plummet 3. AI-native companies take share from legacy incumbents The first one doesn't hold up. The second is too important to dismiss casually, but lacks evidence thus far. The third strikes me as most threatening, but it, too, rests on a flawed logical foundation.   On rolling your own software, customers who choose to build almost always regret it. An internal IT team can build a CRM, but they can't maintain and improve it like a dedicated software company. Anthropic spending 8-figures on Workday for HR software shows that incumbent providers have more to offer than software skeptics realize.   Sure, there's a risk of AI automating away seats (and therefore seat licenses). But this creates opportunity for software incumbents to deliver AI-infused software that allows customers to do just that. Software vendors can easily shift to usage-based or other value-based pricing models. Relatedly, there's a larger fear that AI will trigger massive unemployment among knowledge workers. This deserves to be taken seriously. However, unemployment is currently quite low by historical standards, and current AI is beyond what most considered AGI a few years ago. I think the worriers here are moving the goalposts. The third concern--AI-native companies decimating software incumbents--is the most formidable. 20-25 years ago, Google hoovered up the attention economy by building the dominant consumer Internet product. Could Anthropic or OpenAI do something like this in b2b? It's not my base case, but it's the scenario that goes the farthest with me.   That said, I think there are 2 misunderstandings embedded here.   The first is that only AI startups can use AI tooling to accelerate product velocity. They aren't. The engineers at Shopify and Workday have the same access to Cursor, Claude Code, or open-source AI models.   The second, bigger misunderstanding is that these markets are zero-sum. By definition, technology expands the existing pie. So I disagree with the idea that AI growth will come at the cost of existing software companies. AI will massively expand software spend.   Which brings me to what's hardest to reconcile. If you believe AI-native startups will create major value, you're implicitly underwriting a world with fragmentation and switching costs among knowledge work tools. You can't hold that public software companies are all in the blast radius of Anthropic AND back the Harveys or Open Evidences of the world.   It doesn't add up, as the bull case of one is implicitly the bull case of the other. And while the market is voting like these are opposites, I expect the weighing machine to disagree.

28

8 Comments

EXCLUSIVE: Under the covers of the $18m Thread investment round Harry Brelsford interviews Matt Linn, co-founder and COO of Thread, an AI service desk that automates non-billable resources on the service desk, making them billable and reducing service costs through agentic workflows. Linn shares his background from a 10-year career with an MSP in New York City. Thread is announcing a new round of fundraising at the Series A stage to accelerate R&D and build faster. The company emphasizes its unique brand and customer-centric approach, showcasing this through creative booths at events. For 2026, Thread plans to enhance product offerings with better context for agents, more auto resolutions, and zero-touch tickets.   WATCH THE PODCAST HERE   Matt Linn's Background and Introduction to Thread - Harry Brelsford introduces Matt Linn, co-founder and COO of Thread, a company described as a "Wall Street darling" in the MSP industry. - Matt Linn shares his background, mentioning his 10-year career with an MSP in New York City called RFA, which focused on hedge funds and private equity funds. - Matt explains his passion for solving industry problems and his decision to create Thread to enhance the MSP business model. - Harry Brelsford asks Matt to provide a concise elevator pitch for Thread. Thread's Elevator Pitch and AI Service Desk - Matt Linn describes Thread as an AI service desk that automates non-billable resources on the service desk, making them billable and reducing service costs through agentic workflows. - Thread aims to create a new business model for MSPs with better unit economics and competitiveness at scale. - Harry Brelsford inquires about any breaking news from Thread. - Matt announces a new round of fundraising at the Series A stage, aiming to accelerate R&D to build bigger, better, and faster than ever. Brand Recognition and Innovative Booths - Harry Brelsford comments on Thread's strong brand recognition and the positive feedback they receive. - Matt emphasizes the importance of Thread's radical and customer-centric approach, highlighting their unique and intentional marketing strategies. - Matt shares examples of Thread's creative booths at events, such as a carnival nightmare theme and a Candy Cane Lane booth, to illustrate their commitment to innovation and customer experience. - Harry recalls past experiences with Thread's themed events, including a party bus and various costumes, showcasing their creative marketing efforts. Future Plans and Final Thoughts - Harry Brelsford asks Matt about any upcoming plans for 2026. - Matt reveals that Thread will be investing heavily in product development, focusing on providing better context for agents, more auto resolutions, and zero-touch tickets. - Harry thanks Matt for his participation and mentions the positive feedback on the MSP forecast. - Matt concludes by reiterating Thread's commitment to innovation and improving service efficiency for MSPs. https://tel…

8

1 Comment

The next frontier of AI isn't just about choosing the right model. It's about building moats through implementation. Across the Emergence Capital portfolio, companies like Bland and Federato are scaling faster by embedding Forward-Deployed Engineers. These engineers bridge the gap between AI and reality, ensuring real results in production. Read more from my colleagues Gordon and Kyle on why FDEs are the key to unlocking AI's full value. https://lnkd.in/gcsRRGX2

29

3 Comments