N°01
+6.8pp APY
Net outperformance vs. passive DeFi composite, audited by Armanino LLP over 18 months.
Source: Armanino LLP independent attestation, August 2024.
§1 — Audited Results · Diligence Room
Performance, audited.
A +6.8% APY outperformance against a passive DeFi benchmark, measured by Armanino LLP over an 18-month window from January 2023 through June 2024. This page is the diligence file behind that single number — methodology, period commentary, third-party verification, and the explicit limits of what it does and does not prove.
Past performance does not guarantee future results. PFN Dai Pte. Ltd. is an analytics provider, not a broker-dealer. The figures below describe the audited measurement of historical strategy output and are presented for institutional due-diligence purposes only.
§2 — Methodology & disclosures
How the +6.8% number was built.
The figure on the page above is the arithmetic difference between the audited net APY of the PFN Dai managed strategy and a passive DeFi composite benchmark over the same window. The construction is intentionally narrow. We describe it here so a portfolio manager can replicate the comparison against their own internal book.
Universe
The audited universe consists of 14 production networks — Ethereum, Arbitrum, Base, Optimism, Solana, Aptos, Sui, and seven additional EVM and non-EVM venues indexed by our validator-node infrastructure. Strategies are signal-driven across listed lending markets, liquid-staking derivatives, and curated LP positions surfaced by the Deep Flow v3.2 model. Concentrated, illiquid, or single-counterparty venues were excluded from the audit universe.
Benchmark construction
The benchmark is an equal-weighted composite of passive DeFi positions available on the same networks during the measurement window — ETH staking, major stablecoin lending, and broad-based LP exposure — rebalanced monthly. No active signal, no on-chain intelligence layer, no discretionary risk override. It is the strategy a portfolio would earn by sitting on the same venues and doing nothing.
Fee treatment
All reported figures are net of platform fees, performance fees, and standard on-chain gas. Auditor adjustments for mark-to-market variance, oracle-event windows, and stablecoin depeg episodes are applied symmetrically to strategy and benchmark.
Rebalancing cadence
The strategy rebalances on signal, not on calendar. Median time between rebalances across the audit window was 2.6 trading days. Average signal-to-execution latency measured 168ms across the indexed chains.
Forward-looking statements. References to historical outperformance describe a closed 18-month window and do not project future returns. Strategy composition, fee schedule, and benchmark construction are documented in the full Armanino report, available on request under NDA.
§3 — Evidence
The numbers behind the edge.
Four figures that sit beside the headline APY claim. Read each as a separate line of evidence: latency, coverage, prediction accuracy, and the breadth of the institutional base that has run diligence on the system.
N°02
<180ms
Signal latency across the 14 EVM and non-EVM chains indexed on our validator-node infrastructure.
Source: PFN Dai infrastructure telemetry, Q4 2024.
N°03
0.87AUC
Rug-pull prediction accuracy of the Deep Flow transformer model, evaluated 72 hours pre-event.
Source: Deep Flow v3.2 backtest, 2.1B labeled wallet behaviors.
N°04
184institutional clients
Across 27 countries, including 9 of the top 50 crypto-native funds by AUM.
Source: PFN Dai client roster, January 2026.
§4 — Period commentary
A note on the periods.
A terminal outperformance number hides the path it took to get there. The following chapter notes describe each sub-period of the audit window, the dominant market regime, and how the strategy behaved through it. Read this against your own internal benchmark windows before drawing conclusions.
Chapter 4.1
Q1 2023 — the post-FTX rebuild
The window opens under the overhang of the FTX collapse and a defensive posture across CeFi-linked venues. The strategy under-allocated to bridge and cross-chain lending, capturing the dislocations that followed as counterparties re-rated. Outperformance against the passive composite built steadily through the quarter as the signal layer identified venues tightening risk premia.
The largest single drawdown of the audit window, -11.2%, occurred in this period during a stablecoin depeg event on a third-party venue excluded from the live book the following quarter.
Chapter 4.2
H2 2023 — the LSD unwind
The liquid staking derivative complex compressed as ETH staking yields normalized and incentive programs rolled off. The strategy rotated out of high-tepid LSD loops into curated stablecoin lending venues surfaced by the Deep Flow model. This was the period where the +6.8% delta was mechanically built — a sequence of small rebalances into venues the benchmark did not hold.
Sharpe over this half-window reached 2.41, the highest in the audit.
Chapter 4.3
H1 2024 — recovery and consolidation
The first half of 2024 saw broad-based DeFi recovery and rising TVL across the indexed chains. Outperformance narrowed in absolute basis points as the passive composite caught a rising tide, but the strategy preserved its edge through selective LP exposure on venues where the signal layer maintained conviction. The audit window closes on a stable, repeatable posture rather than a transient spike.
Hit rate of monthly rebalance signals across this period: 64.1%.
§5 — Third-party verification
What third parties have verified.
Four lines of attestation a procurement team should see before booking a diligence call. Each item is named specifically and tied to a public record or counterparty — not a self-claim.
-
01 / Audit
Armanino LLP verified a +6.8% average APY outperformance against passive DeFi benchmarks over the 18-month measurement window ending June 2024. Independent attestation issued August 2024.
-
02 / Security certification
SOC 2 Type II certification achieved February 2023. ISO 27001 certification achieved June 2024. PFN Dai is the only on-chain analytics provider with both certifications and MiCA-aligned EU data residency.
-
03 / Capital backing
Series A closed at $24M in March 2024, led by Polychain Capital with participation from Framework Ventures and a strategic investment by Circle Ventures.
-
04 / Industry recognition
Named Best DeFi Analytics Platform at the HedgeWeek Digital Asset Awards 2024. Shortlisted for the Forbes Fintech 50 — Top 10 AI in Fintech, Q4 2024. Finalist at the 2024 Crypto AI Awards.
SOC 2 · TYPE II ISO 27001 MiCA-ALIGNED · EU ARMANINO LLP · ATTESTED
§6 — Continue diligence
See the platform behind the numbers.
The audit describes what the strategy produced. The platform describes how it produces it. Book a working session with our research team and we will walk a delegated seat through the Deep Flow v3.2 model, the live signal feed, and the chain-by-chain risk surface that drives the figures above.
- 30-minute working session with a member of the research team.
- Read-only access to a sandbox strategy built on your own risk envelope.
- Methodology pack including the full Armanino report under NDA.
Request a Platform Demo
Talk to research one-on-one
Or write directly — [email protected] · +65 6817 4421 · 79 Anson Road, #21-01, Singapore 079906.