Former Cadre Co-Founder Bets on AI to Untangle Private Credit Operations

A Second Act in Alternative Assets

Ryan Williams spent a decade building Cadre, the real estate investment platform he co-founded in 2014, into a venture that attracted over $160 million in funding and reached an $800 million valuation before its acquisition by Yieldstreet in 2024. Now he's applying lessons from that journey to a different layer of the alternative investment stack: the messy, spreadsheet-heavy operational backbone that private credit managers wrestle with daily.

Ellis AI emerged from stealth this week with $10 million in seed funding from a roster that includes First Round Capital, 645 Ventures, Harlem Capital, Khosla Ventures, Thrive Capital, Slow Capital, Kearny Jackson, and Ariel Alternatives CEO Mellody Hobson. The startup targets a problem Williams observed up close during his Cadre years, where modernizing the investor-facing experience did little to address the fragmented systems running underneath.

The Excel Operating System Problem

Private credit firms juggle a patchwork of software tools, accounting systems, and document repositories that rarely talk to each other. Monthly close processes can involve downloading files from multiple platforms, manually reformatting data, hunting down discrepancies, and re-entering information into spreadsheets. For many firms, Excel has become the de facto operating system, a duct-tape solution that scales poorly and invites error.

Ellis AI positions itself as connective tissue rather than replacement infrastructure. The platform integrates with existing systems and document stores, centralizing scattered information into a unified interface. AI agents then layer on top, flagging inconsistencies in the data and automating tasks like portfolio monitoring, report preparation, and month-end reconciliation.

Williams describes the approach as pragmatic. Rather than forcing firms to rip out legacy systems and start fresh, Ellis slots into the existing technology landscape and imposes order on the chaos. The goal is to compress workflows that currently consume days of analyst time into minutes, freeing credit teams to focus on judgment calls rather than data wrangling.

Agents With Guardrails

The architecture keeps humans firmly in the loop. Material decisions and actions remain the domain of human experts, with AI agents handling repetitive, high-volume tasks that don't require discretion. Williams acknowledges that the boundary between machine automation and human oversight will likely shift over time, but he doesn't foresee full autonomy on the horizon.

That position reflects both regulatory reality and client comfort. Private credit managers operate in a compliance-heavy environment where audit trails and accountability matter. Handing over decision-making authority to black-box algorithms would introduce risk that most firms aren't prepared to accept. Ellis's design philosophy centers on augmentation: cut through the noise, surface the signal, and let experienced professionals make the final call.

The startup began development last year, drawing on Williams's firsthand experience with the operational constraints that limit growth in private markets. At Cadre, investor acquisition and platform design received significant attention and resources. Back-office infrastructure, by contrast, remained a persistent bottleneck, cobbled together from point solutions that didn't integrate cleanly.

A Crowded but Immature Category

Ellis enters a fintech landscape where AI-powered workflow automation is attracting significant capital, but where private credit operations remain underserved relative to public markets and traditional banking. The asset class has grown rapidly over the past decade, with dry powder and deal volume both hitting record levels, yet the tooling available to managers has lagged behind the sophistication of the investment strategies themselves.

Competitors in adjacent spaces include platforms focused on fund administration, investor reporting, and compliance automation. Ellis differentiates by targeting the connective layer, aiming to unify disparate systems rather than replace any single function. That approach requires deep integrations and a flexible data model capable of accommodating the idiosyncrasies of individual firms.

The $10 million seed round provides runway to build out the product, onboard early customers, and refine the AI models that power the agent layer. Williams brings a track record and network from his Cadre tenure, assets that should ease customer acquisition in a relationship-driven industry. The presence of Mellody Hobson among the backers signals credibility within the institutional investment community, a constituency Ellis will need to win over to achieve scale.

What Comes After the Seed

Success for Ellis will hinge on execution across multiple dimensions. The product must deliver measurable time savings and error reduction without introducing new points of failure. Integrations need to be robust enough to handle the variety of systems in use across the industry. And the AI agents must prove reliable enough that users trust them with high-stakes workflows.

The private credit market offers a substantial addressable opportunity, but it's also conservative and risk-averse. Firms move slowly when adopting new technology, particularly tools that touch core operational processes. Ellis will need to demonstrate clear ROI and build case studies with early adopters to overcome inertia.

Williams's bet is that the pain of fragmented workflows has reached a tipping point, and that firms are now willing to invest in infrastructure that promises relief. If he's right, Ellis could carve out a defensible position in a category that's ripe for consolidation. If the market isn't ready, or if the product can't deliver on its automation promises, the company will face a longer, harder road to product-market fit.

For now, the focus is on building, onboarding, and proving the concept with a handful of early customers. The seed capital buys time to validate the thesis and refine the approach. Whether Ellis becomes the operating system for private credit or another well-funded experiment will depend on how well it navigates the gap between technical capability and market readiness.