The Venture Codex

Participated · Series A · May 2026

OpsMill builds Infrahub, an infrastructure data management platform that models environments as a network of relationships rather than static assets and is designed as a system of record for IT. Infrahub is built on a graph database with native version control and is offered as a free open-source Community edition alongside a commercial Enterprise version. The platform enables teams to manage dependencies, validate changes, and maintain governance through a structured DevOps workflow. OpsMill targets large infrastructure operators and enterprises across retail, manufacturing, and financial services and has already seen adoption in those sectors. The company emphasizes delivering a unified, trusted data layer to support safe, scalable automation and AI-driven operations. With fresh Series A funding, OpsMill plans to scale the product and expand to meet increasing demand for AI-ready infrastructure data solutions.

Participated · Seed · May 2026

XCaliber Health offers an agentic operating system that connects EHRs and other healthcare systems to orchestrate semi-autonomous workflows with human oversight. The platform delivers pre-built agents for interoperability, analytics, scheduling, prescription refills, referral coordination, and care gap management. XCaliber reports processing more than eight million chart updates and generating over 160,000 EHR updates daily across more than 700,000 unique patients, and its prescription refill agent saves providers an average of six hours of manual work per day. Customers include health systems, provider groups, and digital health partners, and the company emphasizes a human-in-the-loop structure and varying levels of autonomy for different agent functions. With the newly announced seed funding, XCaliber plans to accelerate product development and scale the platform nationwide.

Participated · Series C · Jan 2026

Founded in 2018, Fiddler AI has built an all-in-one observability and security platform that serves as a control plane for compound AI systems, giving enterprises standardized telemetry, evaluation, continuous monitoring, policy enforcement, and auditable governance. The product includes proprietary Trust Models that provide out-of-the-box accuracy, safety, and compliance checks, enabling Fortune 500 customers to diagnose, explain, and manage AI agent behavior in real time. Over the last 18 months the company’s revenue has grown more than 4×, reflecting rapid adoption in regulated sectors such as healthcare, financial services, and insurance. Recent milestones include achieving AWS Pattern Partner status and being ranked the #1 company in AI Agent Security & Risk Management by CB Insights. With its new control-plane capabilities, Fiddler plans to deepen integrations across the AI ecosystem and expand its footprint in mission-critical, agent-driven workflows. The latest financing brings total capital raised to $100 million, supporting continued product development and go-to-market expansion.

Participated · Series A · Oct 2025

Lyzr AI builds AI agents for enterprise use and operates an internal agent called SivaClaw that can engage investors, draft memos, and analyze slide engagement. The company used SivaClaw to run its own capital raise, demonstrating the product by fielding questions from over 130 investors during the process. That fundraising effort resulted in a $100 million Series B at an estimated $500 million valuation and reportedly generated about $400 million in investor interest. Lyzr is roughly three years old and located in Jersey City, New Jersey. The articles do not disclose revenue, user counts, or detailed financial metrics, nor do they list specific participating investors or planned allocation of the new capital.

Participated · Seed · Sep 2025

Founded in 2025 by RapidAPI creator Iddo Gino, San-Francisco-based Datawizz offers a fully managed platform that profiles an organization’s AI requests, trains small Specialized Language Models (SLMs), and automatically routes queries to the most suitable model. These SLMs are roughly 1,000× smaller than typical LLMs yet deliver >85 % cost savings, 5–15× speed improvements, and higher precision when fine-tuned with human feedback. The models are owned and deployed by the customer—on-prem, on-device, or in private clouds—eliminating reliance on external APIs and enhancing data privacy. Datawizz positions its technology as the foundation for a sustainable AI stack and projects that by 2030 about 70 % of AI traffic will run on sub-5 B-parameter models, with over half managed through its platform. The new capital will be used to accelerate product development, expand headcount, and launch go-to-market efforts in the U.S. and Europe. Although still at the seed stage, the company claims transformative unit-economic gains for customers as enterprises struggle with rising LLM API costs.