Participated · Seed · Aug 2026

Synthefy develops foundation models specifically for structured numerical data — data types such as tables, time series, transactions, sensor readings, and other numeric signals. Its flagship open-source model, Nori, is available under an Apache 2.0 license and the company says Nori ranks first across 130 public regression benchmarks; Synthefy reported almost 600K Nori model downloads and about 5,000 Python installations in the weeks after launch. The platform is designed for enterprise deployment with options including an open model, managed API, enterprise platform, and on-premise deployments integrated with AWS, Databricks, Snowflake, and Google Cloud. Synthefy positions its models as smaller, purpose-built architectures that require less compute and energy than large general-purpose language models. Targeted use cases cited include demand forecasting, pricing optimization, fraud and risk detection, infrastructure failure prediction, and operational improvements across retail, financial services, telecommunications, infrastructure, healthcare, and defense. The company was founded in 2023 and the team includes former engineers and researchers from Meta, Nvidia, Adobe, Uber, and top academic institutions.

Participated · Equity · Jul 2026

Pangram develops machine learning models to detect AI-generated and AI-assisted text and images, training on tens of millions of human documents and synthetic mirrors to learn stylistic differences. The company recently launched Pangram 4, its next-generation text detector which it says is over 99% accurate, and Pangram Image, an image detector currently available in research preview. Pangram provides a $20/month web subscription, a Chrome extension that labels content in real time across platforms like X, LinkedIn, Substack, Reddit, and Medium, and an API used by customers such as Substack, Quora, schools, publishers, and recruiters. Founders Max Spero and Bradley Emi, both Stanford AI and machine learning graduates, started the company about two years ago in response to the proliferation of AI-generated content following ChatGPT. The team reports a very low false-positive rate (roughly one in 10,000 human documents flagged) and highlights use cases ranging from publisher integrations to academic and moderation settings. The company has not disclosed revenue figures or detailed financial metrics in the article.

Participated · Equity · Jul 2026

Integral provides Forward Deployed Privacy Services that embed privacy experts directly into customer data pipelines to perform entity-preserving remediation and continuous privacy risk assessment. Its methodology is grounded in peer-reviewed statistical disclosure limitation research and was initially validated in healthcare, life sciences, and health data marketplaces. Integral produces defensibility artifacts such as Expert Determinations under HIPAA or signed defensibility opinions tailored to context, enabling enterprises and data originators to commercialize or share sensitive datasets. The company supports AI labs, data platforms, vertical AI builders, and enterprises operating in regulated environments. Following a $25 million financing, Integral plans to expand its privacy engineering and statistical methodology teams, invest in linkage and continuous assessment infrastructure, and broaden go-to-market efforts across the real-world data economy.

Participated · Series A · Jun 2026

Neion Bio’s core product is the Raptor™ platform, which leverages genetic engineering to restrict recombinant protein production to chicken eggs as a means to manufacture complex, glycosylated biologic proteins. Founded in 2024, the company is positioning the platform to address scalability, resiliency, capital efficiency, speed, and the ability to express difficult-to-manufacture molecules. Neion plans to expand its pipeline across biosimilars, innovative medicines, critical reagents, and animal health and to commercialize via partnerships, including a recently announced multi-product biosimilars co-development and supply agreement with a major pharmaceutical company. The company is advancing proprietary technologies intended to enable production at meaningful scale within days and to shorten iteration cycles for molecule screening and production. Financially, Neion completed an oversubscribed $23 million Series A to fund these platform and pipeline initiatives.

Participated · Series A · Jun 2026

Board makes a 24-inch touchscreen in a wood-finish frame that uses proprietary technology to recognize physical game pieces, combining tactile board-game elements with interactive digital experiences. Founded by Brynn Putnam, the company is three years old and based in New York; Putnam previously sold Mirror to Lululemon for $500 million. Board reports strong early traction, with the device deployed in tens of thousands of homes, schools, hospitals, and restaurants across all 50 states and 85% of customers averaging 30 or more play sessions per month. The company previously raised $15 million in a round led by Lerer Hippeau. As part of its product roadmap, Board plans to launch Board Studio, an AI-powered game-creation platform, later this year.