Participated · Series B · May 2026

OpenRouter operates as middleware between developers and more than 300 open-source and proprietary large language models, offering a unified API that lets developers compare quality, price, and latency and dynamically route tasks to different models. The platform charges a percentage service fee when developers purchase credits to run models, generating revenue that the article reports has grown from about $10 million annualized in October to over $50 million annualized currently. Founded in 2023, the company enables multi-model architectures by allowing applications to call different models for different task types and supports developers in optimizing cost and performance. The article highlights OpenRouter's neutral positioning and broad model coverage as competitive advantages versus cloud providers and startups that may favor their own models. It faces competition from cloud giants offering multi-model access and startups like PortKey, and must manage neutrality and supplier relationships as it scales. The company is positioned as addressing developer demand for simpler multi-model orchestration amid growing AI application complexity.

Participated · Series A · Mar 2026

Knox operates the largest managed federal cloud and a multi-cloud federal boundary across AWS, Azure, and Google Cloud to enable SaaS vendors to obtain FedRAMP authorization rapidly. The company offers a pre-authorized environment that it says reduces first-year FedRAMP costs by about 90% and can achieve authorization in 90 days. Knox holds Authorizations to Operate (ATOs) across 15 federal civilian and defense agencies and serves customers including Adobe, Celonis, OutSystems, Armis, and BigID. Its platform targets both federal and commercial production environments by applying government-grade security and compliance at scale. The company intends to use the Series A proceeds to expand federal authorization capacity and accelerate customer onboarding into authorized environments.

Participated · Series A · Aug 2025

TinyFish develops AI web agents that simulate human navigation to perform repetitive and complex online tasks and to collect data at scale. Its system uses advanced AI models for reasoning and exploration, then encodes that knowledge for deterministic, high-speed execution. The company initially targets retail and travel, with use cases such as dynamic price surveillance—monitoring prices, promotions, shipping times and inventory across competitor sites. TinyFish says its technology helps turn the web into analyzable data to help companies increase revenue rather than just cut costs. Founded in 2024 and based in Palo Alto, California, the company has about 25 employees. It plans to use new funding to invest in product development and expand commercial operations.

Participated · Series C · May 2025

CloudZero provides a next-generation cloud cost optimization platform that automates the collection, allocation, and analysis of cloud costs to uncover savings and improve unit economics. It enables engineering-led cost management so companies can understand 100% of their operational cloud spend and connect cloud decisions directly to business outcomes. Customers named in the announcement include Coinbase, DraftKings, Expedia, Moody's, Nubank, Klaviyo, Miro, and Rapid7. The company reported another year of triple-digit growth and announced a $56 million Series C to fund expansion. CloudZero plans to invest the funding in AI-driven cost analytics and forecasting, deeper integrations across developer and finance workflows, and expansion of strategic partners and cloud alliances. The platform is positioned to help enterprises scale efficiently and improve unit economics as AI workloads reshape infrastructure needs. CloudZero provides engineering teams with a unified dashboard that combines cloud billing data with business- and system-level telemetry to show how much an individual product or feature costs. The platform ingests customers’ cloud, platform-as-a-service and software-as-a-service spend, normalizes it into a common data model, and uses hourly cloud spend data plus an AI algorithm to identify “abnormal” spend events. CloudZero plans to use the new funding to expand its platform and scale go-to-market efforts, with an emphasis on savings insights and self-service analytics. The company has over 100 customers, and CEO Phil Pergola said revenue has increased 10x since 2021, though he would not disclose absolute revenue figures. Co-founded by Erik Peterson and Matt Manger nearly 10 years ago, CloudZero positions itself in the FinOps market alongside players such as ProsperOps, Xonai, Vantage, Cast AI, Zesty, and larger vendors like VMware and Apptio. Pergola said demand for the platform is at an "all-time high" as customers focus on profitability and cloud cost visibility. CloudZero is advancing a SaaS-based site reliability platform that identifies system fragility so engineers can proactively optimize reliability. The platform continuously models the changing surface of a cloud system as it operates under pressure and provides insights that deliver trusted reliability. The product is currently in an invite-only beta and the company also announced an open source project, CloudZero Reactor for AWS, which normalizes AWS service metadata and provides an interface to query that information. The Reactor enables the CloudZero platform to model past, present and future states of systems, helping engineers discover and correct emergent behavior and understand its causes. CloudZero will use the funds to open its headquarters in Boston, grow its enterprise SaaS platform, and increase investment in its open source community. The company is led by founder and CEO Erik Peterson and recently closed a $5M Series A.

Participated · Series A · Dec 2024

Stainless uses AI to generate SDKs from API specs automatically, producing libraries in languages such as Python, TypeScript, Kotlin, and Go. Its platform pushes updates as APIs evolve and supports versioning and changelogs, while generating an initial config teams can fine-tune. Customers include OpenAI, Anthropic, Meta, Runway, Groq, Cerebras, Modern Treasury, and Cloudflare. Stainless reports “hundreds” of paying customers and says its SDKs are downloaded tens of millions of times every week; most customers pay for its enterprise tier. Pricing ranges from a free single-SDK option up to $250/month to $30,000/year for multi-SDK packages. The company’s annual recurring revenue is hovering around $1 million and it is nearing profitability as it pursues a broader API platform vision beyond SDKs. Stainless ingests API specs and produces polished SDKs across languages like Python, TypeScript, Kotlin, Go and Java, while handling versioning and changelogs as APIs evolve. The product uses generative AI to produce an initial configuration that customers can fine‑tune, which the founder says is especially valuable for AI companies with novice developers. Customers include OpenAI, Anthropic, Together AI, Lithic, LangChain, Orb, Modern Treasury and Cloudflare; the company reports “dozens” of paying clients in its beta. Some generated SDKs — notably OpenAI’s Python SDK — are receiving millions of downloads per week. Stainless monetizes via a free single‑SDK publishing tier and a paid enterprise tier (most customers pay enterprise), with pricing ranging from $250/month to $30,000/year. Founder Alex Rattray bootstrapped the business with revenue from day one, and annual recurring revenue is hovering around $1 million; he expects the company could be profitable as soon as this year and plans to expand the product line and headcount.