Matt Bornstein

The original Cursor thesis was interface over models — followed by a later shift for different reasons. In early 2024 the founders were “bitter lesson pilled”: “we don’t need to compete with Anthropic and OpenAI on models right now. The interface between the human and the model is the key thing,” with Michael and Aman saying code would become pseudocode. Matt summarized the product implication as “pairing the programmer’s intent down to the minimum possible spec.” Once Cursor had won users, Sarah said it had the assets, data, and know-how to build its own models — a flip that would have been impossible without starting as a frontier lab.

The Matt-led Series A looked almost irrational against Microsoft. Copilot led “by a mile,” and Microsoft owned VS Code, the OpenAI weights, 100 million developers, and the “greatest enterprise distribution ever” — “they literally own every piece of that.” The a16z team heuristic that carried the bet: always back the leading independent in a large market, even when the incumbent seems very scary.

The growth round required throwing out the growth model itself. Sarah Wang’s team offered to hire a sales leader at the conventional $25–50M ARR self-serve plateau; Michael “looks me dead in the eye” — “self-serve is not petering out” — forcing them to abandon assumptions like growth asymptoting to 25% in five years. The Series A was in May/early summer 2024, and Sarah co-led the B by October on “vertical liftoff”; the company was growing from 4 to 50 in four months “or whatever.”

The founders were “paranoid but unfazed” through a rotating cast of existential competitors — Copilot, Windsurf, Cognition, and Claude Code (May 2025). Matt asked Michael about Claude Code, and Michael replied: “we are going after the biggest market in the world… you’re always going to have formidable competitors. That does not scare us” — what Sarah calls “humility mixed with bravado… the winning combo.” The genuine “oh shit” moments were model releases like Opus 4.5, and Cursor answered by cannibalizing itself “in a Reed Hastings type way”: IDE → agent platform → model platform in two years.

Matt Bornstein dismisses the summer-2025 gross-margin discourse as an X echo chamber. “Tech transformations always precede figuring out the business” — the internet didn’t monetize for years — and investors picking apart nascent transformations have “historically been proven wrong.” The margin answer was enterprise, where labs subsidize self-serve but real margins live in third-party and enterprise, and Cursor executed “maybe the fastest build of a sales team in the history of the planet,” reaching over 50% of the Fortune 500.

Hiring was the hidden machine: founders spending 40% of their time recruiting, AE searches run like research projects. Top 10 companies → top 10 teams → the #1 and #2 individual AEs, then “back-channel, back-channel, back-channel.” Sarah said she had never seen technical product founders spend 40% of their time recruiting. An a16z team sat in the office for eight hours at a stretch helping source.

The endgame combination is, per Sarah, “probably the best fit I’ve ever seen” in M&A, with Matt agreeing: “Elon has the compute, they have the distribution and the data.” Sarah’s parallel is SpaceX in 2019 promising global internet with no Starlink satellites in the sky — and Cursor “beat every forecast they ever gave us… by a lot,” with the arc summed up as competing with Microsoft, then Anthropic, changing product and go-to-market, and undertaking “the most complicated M&A of all time” in two years. The acquirer is unnamed on-air.

🔗 Original source & video: The Company That Made AI Coding Feel Inevitable

vLLM has become a widely used execution layer connecting open-weight models to major accelerators. Introduced as running on half a million GPUs at any moment, it supports more than 1,000 active model architectures while NVIDIA, AMD, Google, Amazon, Intel, and others ensure new chips can run it—and often benchmark against it. Simon Mo likens its role to “databases and operating systems” for AI.

Open weights shifted from enthusiast territory to strategic infrastructure when application companies needed differentiation beyond a proprietary-model wrapper. Matt Bornstein points to Cursor, Decagon, Harvey, and similar startups requiring their own mid-training, post-training, inference, and deployment techniques. Closed APIs do not provide that access, so open source became “deeply embedded,” even though Matt notes OpenAI and Anthropic models remain more widely used and generally more critical overall.

The economic case is increasingly about controllable performance, reliability, and data—not merely cheaper tokens. A voice-agent company can control its infrastructure and enforce a latency SLA. Kimi K2 bridges almost a 10x price gap without being as expensive as Claude or GPT-5, while bringing an Opus 4.1-level model onto infrastructure that users can run and fine-tune. Open-weight providers can potentially offer 10 speed tiers, including 400–500 tokens per second in some workloads, versus a proprietary provider’s regular and fast modes.

Open-weight licensing is moving away from unconditional gifts because frontier training cannot be sustained by donated developer time. Model labs face millions or billions of dollars of compute plus repeated failed runs, leading to usage thresholds, derivative-work provisions, and commercial agreements. Simon’s pharmaceutical analogy captures the requirement: released products must return enough revenue to fund the next risky R&D cycle.

Moderation failures may make open weights the default for trusted, specialized work. Simon argues that proprietary guardrails remain arbitrary and false-positive-prone; even GPU-kernel debugging can trigger restrictions and destroy a two-hour session. “If moderation is never solved,” users will prefer models whose guardrails they can control for trusted use cases.

Simon expects no meaningful open-versus-closed capability gap within one year because progress now depends more on environments and algorithms than distribution strategy. Moonshot’s front-end coding loop—generate, render, inspect, and iterate—is his key example of an environment that cannot simply be distilled. He leans against distillation as the main explanation for progress: the durable engine is “really smart people” combining compute, data, environments, and novel methods.

🔗 Original source & video: How Open Source Became AI’s Backbone | Inferact with a16z

AI is not merely another application wave: the panel treats models as a fourth infrastructure pillar because they change chips, data centers, latency requirements, and the programming model itself. Martin Casado’s dividing line is that applications have “abdicated logic”—rather than programmers encoding every decision, software asks the model to “come up with the answer for me.” For career software people, “software is being disrupted” and starting to eat itself.

The supercycle thesis is that cheaper capabilities expand the TAM, create new users and behaviors, and leave incumbents poorly equipped for the resulting white space. The panel’s blunt investor heuristic is that “infra creates TAM”; dismissing a developer tool because today’s market looks small risks missing another GitHub. Natural language also delivers the low-code promise by making programming accessible to anyone with domain knowledge and an idea.

Developer distribution is becoming consumer-like just as the technical audience grows from the low tens of millions to above 50 million. Individual developers increasingly discover and adopt tools bottom-up, while companies still evaluate them through centralized technical buying centers. The panel therefore emphasizes understanding both individual adoption and enterprise buying centers.

The panel has moved away from its early thesis that AI offered “no defensibility anywhere in the stack.” In today’s expansion phase, zero-sum thinking is “deadly”: chips, clouds, models, and applications can all grow simultaneously, while deep engineering, distribution, embedded integration logic, and high switching costs preserve value. Consolidation may eventually produce oligopolies or monopolies, but infrastructure layers rarely disappear.

The panel highlights a shift from prompt engineering to context engineering: model performance depends on selecting the right data, tools, priorities, and guarantees before each call. That creates potential infrastructure around data pipelines, indexes, prioritization, and formal guarantees. The panel expects a new software formalism to emerge over roughly five years, not a world where natural-language wishes eliminate systems engineering.

Agents work best where their loops contain objective error correction, which is why coding is ahead of general web automation. Code can be linted, interpreted, compiled, and tested, arresting error propagation; a vaguely instructed agent sent to “wander out in the woods and bring back a bear” still breaks down. Martin has nevertheless become a convert for bite-sized, well-articulated coding tasks.

Better coding tools are more likely to create more developers and software than to collapse engineering employment. Programming remains creative specification, while customers buy software because someone encoded the correct workflow and domain decisions—not because CRUD applications are inherently difficult to type. The median pull request reportedly changes only two lines, underscoring that understanding the need is often harder than implementing it.

🔗 Original source & video: The Future of Software Development - Vibe Coding, Prompt Engineering & AI Assistants