Ellis
by Ellis
AI-native operations platform that reconciles private credit fund data and runs the close, reporting and monitoring
Ellis is an AI-native operations platform for private credit managers. It connects to the fund administrator, loan servicing system, general ledger, bank feeds and Excel models a firm already runs, reconciles them into one traceable book, and deploys agents for month-end close, LP reporting, cash forecasting, covenant monitoring and exception handling. Built for direct lenders and multi-strategy credit firms.
Ellis is an AI-native operations platform for private credit managers, built around a unified data layer that reconciles the systems a fund already runs rather than replacing them. It connects directly to fund administrators, loan-servicing platforms, general ledgers, bank feeds, portfolio data, borrower financials, loan tapes, compliance certificates and regulatory filings, captures each source as-is with no rekeying, and produces a single reconciled book in which every figure traces back to the originating general ledger account or spreadsheet cell. On top of that layer it runs purpose-built agents for reconciliation and break resolution, month-end close, LP and regulatory reporting, cash-flow forecasting, covenant and portfolio monitoring, anomaly detection, exception tracing, SBIC compliance and document intelligence. Users query live portfolio data in natural language and generate evidence-linked reports with source-level audit trails, and founder Ryan Williams is explicit that material decisions and actions remain with human experts rather than the agents. The product targets a gap the industry's own data describes bluntly: Allvue's 2026 GP outlook survey found 87% of firms already use AI for investment due diligence and document review but only 3% apply it to reconciliation, even as 70% of operations teams name manual work and Excel dependence as their top technology problem. Ellis was founded in 2025 and is based in New York by Ryan A. Williams, who co-created the real estate investment platform Cadre in 2014, with former WeWork and Wonder product leader Jason Liao as chief product officer. It emerged from stealth on 30 July 2026 with more than $10 million in seed funding led by First Round Capital, and its named users and design partners — including Treville Capital Group, Lafayette Square and Kennedy Lewis — sit within a partner network representing over $50 billion in cumulative assets under management.
The COO or head of fund operations at a direct lender or multi-strategy private credit manager whose month-end close still runs on spreadsheet reconciliation between the fund administrator, the loan servicer and the general ledger.
One reconciled book where every reported figure traces back to its source GL account or cell, so close, LP reporting and covenant monitoring run on agents instead of manual tie-outs.
At a Glance
- Category
- Data & Analytics
- Pricing
- Contact for pricing, Subscription
- Target Market
- COOs, CFOs, Fund Operations Leaders, Private Credit Managers, Controllers, Compliance Officers
- Deployment
- Cloud-only
- Founded
- 2025
- Headquarters
- New York City, United States
- Customers
- Design-partner network representing $50B+ in cumulative AUM, including Treville Capital Group, Lafayette Square and Kennedy Lewis
Key Features
- ✓Unified reconciled data layer
Connects fund admin, loan servicing, general ledger, bank feeds and spreadsheet models and captures each source as-is, without rekeying.
- ✓Source-level traceability
Every figure in a report links back to the originating GL account or spreadsheet cell, which is what makes the output defensible in audit.
- ✓Reconciliation agent with proposed fixes
Surfaces breaks between systems and proposes remediations rather than only flagging that the numbers disagree.
- ✓Close and reporting agents
Draft month-end close packages and LP reports automatically, addressing the manual tie-out work that dominates fund operations calendars.
- ✓Cash forecasting and scenario analysis
Projects fund and portfolio cash flows and runs scenarios against live reconciled data instead of a stale spreadsheet snapshot.
- ✓Covenant and portfolio monitoring
Tracks borrower compliance certificates and financials continuously, flagging covenant pressure and anomalies as data arrives.
- ✓Natural language querying
Lets operations and deal staff interrogate live portfolio data conversationally and generate evidence-linked outputs from the answer.
Capabilities
Use Cases
- •Automating month-end fund close
An operations team replaces multi-day manual tie-outs between fund administrator and general ledger with agent-run reconciliation and a drafted close package.
- •Evidence-linked LP reporting
A manager produces investor reports where every number carries an audit trail to its source, shortening the LP diligence and fundraising cycle.
- •Covenant and borrower monitoring at scale
Compliance certificates and borrower financials are ingested continuously so covenant breaches surface early rather than at quarter end.
- •Reconciliation break investigation
When the servicer and the ledger disagree, the exception-tracing agent isolates the discrepancy and proposes a fix instead of a manual hunt.
- •Cash forecasting for capital calls and distributions
Treasury runs forward-looking scenarios on reconciled live data to time capital calls and distributions with more confidence.
Ideal For
Best For
- ✓Private credit managers reconciling fund administrator, loan servicing and general ledger data every month
- ✓Fund operations teams automating month-end close and exception resolution
- ✓LP and regulatory reporting where each figure must carry a source-level audit trail
- ✓Covenant and cash-flow monitoring across a direct lending portfolio
- ✓SBIC compliance workflows and evidence-linked regulatory filings
- ✓Multi-strategy credit firms trying to replace Excel-based portfolio monitoring
Not Ideal For
- ✗Firms outside private credit today — the agents, data model and compliance workflows are purpose-built for credit funds, with broader alternatives coverage stated as a future plan rather than shipped
- ✗Institutions that require a completed SOC 2 Type II report before onboarding a vendor, since Ellis states its Type II audit is still in progress
- ✗Small funds without meaningful system sprawl, where the integration lift across fund admin, GL, servicing and bank feeds outweighs the reconciliation savings
- ✗Buyers who want a production-proven platform with independent references, given the company launched publicly in July 2026 and is still design-partner stage
Deployment
Market & Ratings
Design-partner network representing $50B+ in cumulative AUM, including Treville Capital Group, Lafayette Square and Kennedy Lewis
Market Analysis
Pros
- ✓Attacks a genuinely underserved workflow — reconciliation — which industry survey data shows is the least automated part of private credit operations despite being the biggest manual burden
- ✓Source-level traceability is the right architectural answer to why AI has struggled in audited finance workflows: outputs are checkable, not just plausible
- ✓Overlay rather than rip-and-replace materially lowers adoption risk for a fund that cannot switch administrators mid-life
- ✓Credible named design partners and a partner network representing over $50 billion in cumulative AUM at launch, which is unusual for a company emerging from stealth
- ✓Strong investor syndicate — First Round leading with Khosla, Thrive, 645 Ventures, Harlem Capital and Mellody Hobson participating
Cons
- ✗SOC 2 Type II is stated as in progress, not complete. For institutional credit managers this is often a hard gate in vendor diligence, and it is the single most likely blocker on a deal today
- ✗The company launched publicly on 30 July 2026 and is design-partner stage; claims of better margins, investor relations and faster fundraising are vendor-reported with no independent benchmark behind them
- ✗No presence on any independent review surface — G2, Capterra and Hacker News returned nothing — so there is no unfiltered production experience to read before buying
- ✗The founder's prior company, Cadre, raised over $160 million and peaked near an $800 million valuation before selling to Yieldstreet in 2024, a track record that cuts both ways in diligence
- ✗Connecting fund administrator, general ledger, loan servicing and bank feeds is a substantial integration project, and the platform's value depends entirely on the completeness of those connections
- ✗No public API is documented, which limits how far a firm can extend the platform into its own internal tooling
- ✗Scope is private credit only today; broader alternatives coverage is stated as a plan rather than shipped capability
Pricing
Enterprise
Contact for pricing
- ✓Unified data layer and connectors
- ✓Reconciliation, close and reporting agents
- ✓Cash forecasting and scenario analysis
- ✓Portfolio and covenant monitoring
- ✓Source-level audit trails
- ✓Per-firm logical isolation and MFA
No pricing is published on the Ellis site and none appeared in the launch coverage; the company emerged from stealth in July 2026 and is selling directly to private credit managers through design-partner style engagements. Expect enterprise contracts scoped to fund count, AUM or connector footprint, negotiated case by case, and expect the integration work across fund administrator, general ledger, servicing and bank feeds to carry its own implementation cost.
Security & Compliance
Sources
This page was written from 7 sources, 5 on domains other than ellis.ai.
- 1.techcrunch.com — repeat founder ryan williams raises 10m seed for an ai start
- 2.businesswire.com — Ellis Emerges From Stealth With More Than $10 Million to Bui
- 3.lasvegassun.com — ellis emerges from stealth with more than 10 milli
- 4.businesswire.com — Allvue Systems Introduces Intelligent Loan Operations Closin
- 5.alternativecreditinvestor.com — allvue systems launches private credit benchmarks based on d
- 6.ellis.ai — ellis.aivendor
- 7.ellis.ai — ellis emerges from stealthvendor
Stay Ahead of the Curve
Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.
SubscribeRelated Products
Monte Carlo
Data and AI agent observability for enterprises running production AI
DataBahn
Agentic data control plane that cuts security telemetry 40-70% before it reaches your SIEM
Actian VectorAI DB
A local-first vector database for AI that runs on the edge, on-prem and air-gapped
Nimble
Expert-level web search agents that learn your domain, cutting AI research token spend roughly in half