Delphina Raises $7.5M to Scale Its AI Analytics Context Factory

Delphina announced $7.5 million in new funding on November 12, 2026, bringing its total funding to $15 million as it launched the Context Factory, a system that uses AI agents to build and maintain the business context behind analytics answers.

Radical Ventures and Costanoa co-led the financing, with participation from Tenacity Capital and angel investors including Fei-Fei Li, Nobel laureate Guido Imbens and early employees of Anthropic and OpenAI. The stated uses of proceeds are scaling the Context Factory and expanding Delphina’s go-to-market and customer-engineering teams.

The new product became available with the announcement. Its central premise is that accurate AI analytics requires knowledge of which tables are trusted, how metrics are defined and what experienced analysts know but have not documented. Delphina says companies usually assemble that context by hand, leaving it to become stale as the business changes.

Agents Maintain Business Context

The Context Factory assigns that maintenance work to agents. Delphina says they can assemble an initial baseline from existing company data and documents in hours, identify and repair gaps as employees ask questions, and route consequential changes to domain experts for approval. That arrangement leaves experts responsible for definitions while agents handle the search and updating work.

Delphina’s product overview says the system reads dbt projects, dashboards, documents and query logs, then continues reading those sources as they change. Approved changes are stored in a versioned knowledge base owned by the customer, and evaluation tests can draw on trusted dashboards that employees have already verified.

When a question exposes a missing metric definition, an illustrated workflow shows the agent searching dbt, Looker and Notion, identifying the owner of the relevant data table and requesting confirmation in Slack. After approval, it publishes the definition, regenerates the original answer and saves the fix as a regression test so the approved logic cannot silently regress.

For enterprise deployment, Delphina says connections use per-user, read-only credentials and each customer receives a separate data plane. Python and DuckDB run in an isolated Firecracker microVM with no internet access, while prompts, queries and knowledge references are logged and can be exported. Those controls make agent activity inspectable while restricting its access to company systems and external destinations.

Adoption Beyond Data Teams

Delphina reports that business teams now ask 73% of the questions on its platform and that CEOs are active users at more than half of its customers. At Substack, more than 70% of employees use the product, including CEO Chris Best. Other customers include LATAM Airlines, Upstart, Udemy and Coursera.

Best described the effect on Substack’s decision-making: “Now our teams get accurate answers in seconds instead of waiting weeks in a queue. It’s changed how we make decisions.”

CEO and cofounder Jeremy Hermann previously built and led Uber’s Michelangelo machine-learning platform and cofounded Tecton, which Databricks acquired. President and cofounder Duncan Gilchrist led global pricing data science at Uber. Gilchrist said operations, sales, marketing and finance teams are now Delphina’s heaviest users.

The funding gives Delphina resources to expand the product and the teams selling and supporting it. The central test is whether adoption by business users and CEOs can grow without weakening the expert approvals, versioned definitions and regression tests on which its promise of trusted analytics depends.