News
Entertainment
Science & Technology
Life
Culture & Art
Hobbies
News
Entertainment
Science & Technology
Culture & Art
Hobbies
The Second-Order Effects of AI | Tomasz Tunguz
AI vendor revenue will double classic software in terms of new bookings this year. This trend is so large it’s starting to have second-order effects. MongoDB reported strong Q2 FY'26 results, delivering $591M in revenue with 24% year-over-year growth. AI is causing a second-order effect & a resurgence in growth in Atlas, the cloud-hosted version of MongoDB, which represents 74% of total revenue. We’ve seen a reacceleration within the hyperscalers already, but now the impacts are felt beyond.
Netflix's New AI Role | Tomasz Tunguz
Netflix invented a new role for their data team : the Media ML Data Engineer. Unstructured data is fundamentally different. It’s multimodal & contains derived fields like embeddings, captions, & transcriptions. It’s also at least 80% of the world’s data & essential for the field of AI. This new role highlights how one of the most important companies within the data ecosystem has evolved to promote multimodal data as core. Software engineering & data engineering are fusing.
What Will You Reinvent Today? | Tomasz Tunguz
The familiar rhythms of work are shifting beneath our feet. For the past fifteen years, we’ve lived in a strange plateau, standardizing workflows & settling into predictable patterns. The tools remained largely the same, the processes became routine, & many of us found comfort in that stability. But something fundamental has changed. Many of these processes can be automated, improved, & reimagined. It’s hard for anyone who has been trained for years to work one way & then be asked to do it differently.
Achieving Flow with AI | Tomasz Tunguz
On September 4 at 5-9pm PDT in Berkeley, Hamel Husain will be leading a conversation featuring Claire Vo, Greg Ceccarelli, and me talking about how to achieve Mihaly Csikszentmihalyi’s flow state with AI. Hamel Husain is a machine learning engineer with over 20 years of experience. He has worked with companies such as Airbnb & GitHub, which included early LLM research used by OpenAI for code understanding. He has also led & contributed to numerous popular open-source machine-learning tools & is currently an independent consultant helping companies build AI products.
When One AI Grades Another's Work | Tomasz Tunguz
Since launching EvoBlog internally, I’ve wanted to improve it. One way of doing this is having an LLM judge the best posts rather than a static scoring system. I appointed Gemini 2.5 to be that judge. This post is a result. The initial system relied on a fixed scoring algorithm. It counted words, checked readability scores, & applied rigid style guidelines, which worked for basic quality control but missed the nuanced aspects of good writing.
Who is Talking to My Users? by @ttunguz
In 1999, the dotcoms were valued on traffic. IPO metrics revolved around eyeballs. Then Google launched AdWords, an ad model predicated on clicks, & built a $273b business in 2024. But that might all be about to change : Pew Research’s July 2025 study reveals users click just 8% of search results with AI summaries, versus 15% without - a 47% reduction. Only 1% click through from within AI summaries.
From Knowledge to Action by @ttunguz
GPT-5 launched yesterday. 94.6% on AIME 2025. 74.9% on SWE-bench. As we approach the upper bounds of these benchmarks, they die. What makes GPT-5 and the next generation of models revolutionary isn’t their knowledge. It’s knowing how to act. For GPT-5 this happens at two levels. First, deciding which model to use. But second, and more importantly, through tool calling. We’ve been living in an era where LLMs mastered knowledge retrieval & reassembly.
Small Action Models Are the Future of AI Agents by @ttunguz
2025 is the year of agents, & the key capability of agents is calling tools. When using Claude Code, I can tell the AI to sift through a newsletter, find all the links to startups, verify they exist in our CRM, with a single command. This might involve two or three different tools being called. But here’s the problem: using a large foundation model for this is expensive, often rate-limited, & overpowered for a selection task.
The AI Revenue Rocket Ship - What 5x Growth in Six Months Means for Everyone by @ttunguz
The AI industry just posted numbers that would make even the most aggressive growth hackers dizzy. How do you grow revenue 5x in six months when you’re already at billion-dollar scale? For most technology companies, sustaining triple-digit growth becomes mathematically impossible once you reach meaningful scale. The law of large numbers kicks in. Customer acquisition costs rise. Market saturation looms. Yet Anthropic defied these constraints, rocketing from a $1 billion annual run rate to $5 billion between Q1 and Q2 2025.
The AI-Driven Cloud Market Share Shift by @ttunguz
What force could dethrone AWS after more than a decade of unchallenged dominance? For years, Amazon Web Services ruled the cloud infrastructure market. It was the default choice without a question for every startup. Then OpenAI released GPT-4. Microsoft’s exclusive partnership with OpenAI transformed Azure from a second-place player into the obvious choice for AI-first companies. With this week’s earnings, we are seeing the ultimate impact of that strategic decision.
The Question I Ask Myself Before I AI by @ttunguz
In working with AI, I’m stopping before typing anything into the box to ask myself a question : what do I expect from the AI? 2x2 to the rescue! Which box am I in? On one axis, how much context I provide : not very much to quite a bit. On the other, whether I should watch the AI or let it run. If I provide very little information & let the system run : ‘research Forward Deployed Engineer trends,’ I get throwaway results: broad overviews without relevant detail.
Hidden Technical Debt in AI by @ttunguz
That little black box in the middle is machine learning code. I remember reading Google’s 2015 Hidden Technical Debt in ML paper & thinking how little of a machine learning application was actual machine learning. The vast majority was infrastructure, data management, & operational complexity. With the dawn of AI, it seemed large language models would subsume these boxes. The promise was simplicity : drop in an LLM & watch it handle everything from customer service to code generation.
The Sales Strategy Conquering the AI Market by @ttunguz
What happens when technology evolves faster than your sales process can adapt? The last fifteen years, startups focused on building software around very well understood processes. We had built an assembly line for software sales, SDR to AE to customer success manager. We calculated ratios between these three total cost of sales and drove the factory to ever improved yields. AI is upending all of that. The underlying workflows are changing so quickly, software buyers no longer know what the ideal processes are, much less which is the best software to buy.
PageRank in the Age of AI by @ttunguz
The internet is about to look a whole lot more like the online advertising world. No, I don’t mean there’ll be more ads. In fact, I think there’ll be far fewer. But the technology stack for content distribution will mirror the architecture that has been implemented in the online ad world. As we’ve reached the AI search tipping point, publishers face an existential challenge : ensuring AI systems use their content in answers to maintain relevancy.
Why Seed Rounds Are Growing as Startups Shrink by @ttunguz
Why is the sub-$5 million seed round shrinking? A decade ago, these smaller rounds formed the backbone of startup financing, comprising over 70% of all seed deals. Today, PitchBook data reveals that figure has plummeted to less than half. The numbers tell a stark story. Sub-$5M deals declined from 62.5% in 2015 to 33% in 2024. This 29.5 percentage point drop fundamentally reshaped how startups raise their first institutional capital.
The AI Search Tipping Point by @ttunguz
OpenAI receives on average 1 query per American per day. Google receives about 4 queries per American per day. Since then 50% of Google search queries have AI Overviews, this means at least 60% of US searches are now AI. It’s taken a bit longer than I expected for this to happen. In 2024, I predicted that 50% of consumer search would be AI-enabled. But AI has arrived in search.
The Rise of the Agent Manager by @ttunguz
If 2025 is the year of agents, then 2026 will surely belong to agent managers. Agent managers are people who can manage teams of AI agents. How many can one person successfully manage? I can barely manage 4 AI agents at once. They ask for clarification, request permission, issue web searches—all requiring my attention. Sometimes a task takes 30 seconds. Other times, 30 minutes. I lose track of which agent is doing what & half the work gets thrown away because they misinterpret instructions.
Budgeting for AI in Your Startup by @ttunguz
For the last decade, the biggest line item in any startup’s R&D budget was predictable talent. But AI is pushing its way onto the P&L. How much should a startup spend on AI as a percentage of its research and development spend? 10%? 30%? 60? There are three factors to consider. First, the average salary for a software engineer in Silicon Valley. Second is the total cost of AI used by that engineer.
The Hungry, Hungry AI Model by @ttunguz
When you query AI, it gathers relevant information to answer you. But, how much information does the model need? Conversations with practitioners revealed the their intuition : the input was ~20x larger than the output. But my experiments with Gemini tool command line interface, which outputs detailed token statistics, revealed its much higher. 300x on average & up to 4000x. Here’s why this high input-to-output ratio matters for anyone building with AI:
Revenue Comes to Crypto by @ttunguz
If I have a dollar to invest in a stock or a crypto token, how do I decide? I need to compare across the two. Historically, that comparison was impossible. Crypto traded on a potent cocktail of hype, narrative, & the promise of a decentralized future. Perception drove valuations. That’s changing. The word “revenue” is no longer verboten in the world of crypto. It’s becoming the goal. This trend will unlock the next wave of institutional capital because investors can compare the risk/reward of crypto with the same metrics as other software companies.
Figma's S-1: A PLG Powerhouse by @ttunguz
Yesterday, Figma filed its beautifully designed S-1. It reveals a product-led growth (PLG) business with a remarkable trajectory. Figma’s collaborative design tool platform disrupted the design market long-dominated by Adobe. Here’s how the two companies stack up on key metrics for their most recent fiscal year: Metric (2024) Figma Adobe Revenue (YoY Growth) $749M (48%) $21.5B (11%) Gross Margin 88.3% 89.0% Non-GAAP Op Margin 17.0% 44.5% Sales Efficiency 1.00 0.39 Adjusted FCF Margin1 24.
Voice, Context & Control: The Three Pillars of Useful AI Email by @ttunguz
Gmail’s AI email assistant writes like a committee of lawyers designed it. Pete Koomen’s recent post Horseless Carriages explains why: developers control the AI prompts instead of users. In his post he argues that software developers should expose the prompts and the user should be able to control it. He inspired me to build my own. I want a system that’s fast, accounts for historical context, & runs locally (because I don’t want my emails to be sent to other servers), & accepts guidance from a locally running voice model.
Why Data is More Valuable than Code by @ttunguz
In “Data Rules Everything Around Me,” Matt Slotnick wrote about the difference between SaaS & AI apps. A typical SaaS app has a workflow layer, a middleware/connectivity layer, & a data layer/database. So does an AI app. AI makes writing frontends trivial, so in the three-layer cake of workflow software the data matters much more. The big differences between an AI & the SaaS app lie within the ganache of the middle layer.
The Multimodal Lake House : Partnering with Lance by @ttunguz
Remember when you took a family photo & Ghibli-styled it? Or that vibe coding session, when you pasted a screenshot of the browser so the AI can help you debug some Javascript? Today, we expect AI to be able to hear, see, & read. This is why multimodal is the future of AI. Multimodal data means using text, images, video, sound, even three-dimensional shapes with AI. These are magical user experiences.
Fighting for Context by @ttunguz
Systems of record are recognizing they cannot “take their survival for granted.” One strategy is to acquire : the rationale Salesforce gives for the Informatica acquisition. Another strategy is more defensive - hampering access to the data within the systems of record (SOR). Unlike the previous software era where SORs built platforms on top of themselves to develop broader ecosystems (in Salesforce’s case Veeva & Vlocity), the AI shift does seem to be more defensive.
DuckLake - Subsecond Latency on a Petabyte by @ttunguz
DuckLake is one of the most exciting technologies in data. While data lakes are powerful, the formats that manage them have become notoriously difficult to work with. “I think one of the things in DuckLake that we managed to do is to cut, I want to say like 15 technologies out of this stack.” How does it achieve this? Instead of building a custom catalog server, DuckLake uses a simple, elegant idea: a standard database to manage metadata.
The Data Decacorn Derby by @ttunguz
Databricks seems to be closing the gap on Snowflake faster than expected. Last week Databricks shared some important updates on their business which allows us to compare the progress of the two companies. Quarterly revenue between the two company shows nearly identical slope, two parallel lines. Snowflake recently exceeded $1b in quarterly revenue mark while Databricks just touched $750m and is targeting $925m for the next quarter. Snowflake’s revenue growth rate has been on a long glide path to nearly asymptoting at 25% year over year.
The Coming Wave of Acquihires by @ttunguz
The Seed Surge of 2021 will lead to a raft of acquihires. In 2021 the total number of US software & AI seeds jumped from 2900 to 4300 - a 49% jump. Seeds fell to about 3000 creating a seed tabletop. Series As moved in lockstep both on the way up and the way down - creating a squeeze. These data form part of a longer term trend of a greater number of seeds but a relatively fixed number of Series As.
Partnering with Maze Security by @ttunguz
Doctors and security research have more in common than you might think. Doctors defend human bodies against an ever-shifting landscape of viruses & infections. Security researchers do the same thing, but at massive scale—protecting thousands of servers instead of a single patient. The doctors’ responsibility are to defend a human body from an ever-shifting landscape of potential viruses and infections. Each human body is slightly different. The research around human health evolves all of the time as well as the research around potential infections.
Patterns Across 5 Years of YC Investing by @ttunguz
Where venture capital flows, innovation follows. And for more than a decade, few faucets have been watched more closely than Y Combinator. An analysis of their investment patterns since 2020 doesn’t just reveal the accelerator’s strategy—it provides a map to the entire startup ecosystem’s next chapter. With Demo Day approaching this week & inspired by Jamesin Seidel’s YC Series A analysis, I wondered how YC investment patterns have changed since 2020.
The IPO Door is Swinging Open by @ttunguz
Who could have predicted that crypto and data center real estate would be the categories swinging the IPO market doors open? In late 2024, I predicted a thaw in the IPO market. We’re now seeing that forecast come to life with CoreWeave and Circle’s IPOs. Neither company is pure-play software, but their strong performances signal renewed investor appetite for the ragged edge of technology. CoreWeave went public in March 2025, raising $1.
Stuck in the Middle of AI Workflows by @ttunguz
Whenever I hear about a new startup, I pull out my research playbook. First, I understand the pitch, then find backgrounds of the team, & tally the total raised.1 Over the weekend, I decided to migrate this workflow to use AI tools, & the process taught me something important about how we’re actually integrating AI into our work. Tools are small programs that expand AI capabilities. ChatGPT might call a web search tool to read a blog post I’d like to summarized.