dS = μS dt + σS dW    VaR_α = −inf{x : P(L > x) ≤ 1−α}    Σ = E[(r−μ)(r−μ)ᵀ]    ΔP ≈ δΔS + ½γ(ΔS)² + θΔt    HHI = Σᵢ sᵢ² E[R_p] = Σ wᵢE[Rᵢ]    σ_p² = wᵀΣw    SR = (μ−r_f)/σ    CVaR_α = E[L | L > VaR_α]    β = Cov(rᵢ,r_m)/Var(r_m)    N_eff = (Σwᵢ)²/Σwᵢ² ∂V/∂t + ½σ²S²∂²V/∂S² + rS∂V/∂S − rV = 0    SaR(α) = Q_α(slippage)    P(default) = Φ(−DD)    λ(t) = λ₀ exp(αN_t + βΣe^{−δ(t−tᵢ)}) max_w {wᵀμ − γwᵀΣw}    r_t = μ + σ_t ε_t    σ²_t = ω + αr²_{t−1} + βσ²_{t−1}    ESaR(α) = E[S | S > SaR(α)]    CR₁ = max(wᵢ)/Σwᵢ ρ(X,Y) = Cov(X,Y)/(σ_X σ_Y)    HF = (C × LT)/D    IF ≥ TSaR(α) × (1+κ)    L = Σ max(0, Dᵢ − Cᵢ × LT)    Δ = ∂V/∂S    Γ = ∂²V/∂S² f(r) = (2π)^{−n/2}|Σ|^{−½} exp(−½(r−μ)ᵀΣ⁻¹(r−μ))    κ = E[(X−μ)⁴]/σ⁴ − 3    ES = −(1/(1−α))∫_α¹ VaR_u du    dV = δdS + ½Γ(dS)² R(t) = ln(S_t/S_{t−1})    E[max(S−K,0)] = SΦ(d₁) − Ke^{−rT}Φ(d₂)    MDD = max_{t∈[0,T]}(M_t − S_t)/M_t    θ = −∂V/∂T    ν = ∂V/∂σ P(cascade) = 1 − ∏(1−pᵢ)    funding = (mark−index)/8h    slippage(q) = ∫₀ᵍ (P(x)−P₀)dx / q    OI_cap = f(σ,depth,IF)    ADL_trigger = IF < threshold w* = Σ⁻¹μ / 1ᵀΣ⁻¹μ    TE = σ(r_p − r_b)    IR = α/TE    J_t = ΣN(0,δ²)    corr_stress = ρ₀ + (1−ρ₀)(1−e^{−λΔVol})    RR = E[recovery|default] V(S,t) = e^{−r(T−t)} E_Q[payoff]    d₁ = [ln(S/K)+(r+σ²/2)T]/(σ√T)    LGD = 1 − RR    PD × LGD × EAD = EL    u·∇u + ∇p = ν∇²u σ_implied = BSM⁻¹(V_market)    Skew = E[(r−μ)³]/σ³    VIX = 100√(2/T Σ ΔKᵢ/Kᵢ² e^{rT} Q(Kᵢ))    Ω(r) = E[max(r−τ,0)]/E[max(τ−r,0)] dS = μS dt + σS dW    VaR_α = −inf{x : P(L > x) ≤ 1−α}    Σ = E[(r−μ)(r−μ)ᵀ]    ΔP ≈ δΔS + ½γ(ΔS)² + θΔt    HHI = Σᵢ sᵢ² E[R_p] = Σ wᵢE[Rᵢ]    σ_p² = wᵀΣw    SR = (μ−r_f)/σ    CVaR_α = E[L | L > VaR_α]    β = Cov(rᵢ,r_m)/Var(r_m)    N_eff = (Σwᵢ)²/Σwᵢ² ∂V/∂t + ½σ²S²∂²V/∂S² + rS∂V/∂S − rV = 0    SaR(α) = Q_α(slippage)    P(default) = Φ(−DD)    λ(t) = λ₀ exp(αN_t + βΣe^{−δ(t−tᵢ)}) max_w {wᵀμ − γwᵀΣw}    r_t = μ + σ_t ε_t    σ²_t = ω + αr²_{t−1} + βσ²_{t−1}    ESaR(α) = E[S | S > SaR(α)]    CR₁ = max(wᵢ)/Σwᵢ ρ(X,Y) = Cov(X,Y)/(σ_X σ_Y)    HF = (C × LT)/D    IF ≥ TSaR(α) × (1+κ)    L = Σ max(0, Dᵢ − Cᵢ × LT)    Δ = ∂V/∂S    Γ = ∂²V/∂S²
Home About Research
Quantitative research and risk advisory. DeFi protocol risk, systematic trading systems, and advanced quantitative analytics.
Projects & Services
CalcMonte
Live
Financial calculator combining loan amortization analytics with Monte Carlo investment simulation. Single-file app, no sign-up, runs entirely in your browser.
calcmonte.com →
Quantitative Risk & Research
Consulting
Quantitative risk analysis for DeFi protocols and financial platforms. Correlated risk metrics, liquidation modeling, insurance fund optimization, incentive design, and portfolio construction. Formerly Gauntlet. Deep expertise in on-chain data pipelines, order book microstructure, and stress testing under tail-risk scenarios.
quant@sepperlabs.com
Contact
General contact@sepperlabs.com
Quant & Risk quant@sepperlabs.com
About

SepperLabs is a quantitative research and risk advisory practice led by Otar Sepper, formerly of Gauntlet. The firm focuses on DeFi protocol risk, systematic trading systems, and quantitative analytics across decentralized and traditional markets.

The practice operates at the intersection of quantitative finance, complex systems modeling, and market design — working with DeFi protocols, digital asset funds, and investment teams seeking rigorous, first-principles analysis of risk, solvency, and capital efficiency.

Quantitative risk is the core focus, bringing deep experience in margin parameter calibration, liquidation mechanism design, and insurance fund sizing — grounded in statistical rigor, simulation-based stress testing, and reproducible analytical frameworks.

While the practice maintains deep expertise in decentralized finance infrastructure, its foundation is rooted in classical quantitative finance: systematic strategy development, portfolio optimization, risk decomposition, and statistical modeling across equities, futures, and commodities markets.

Core Focus Areas
DeFi Protocol Risk & On-Chain Market Design
  • Protocol-level risk architecture for lending and derivatives platforms
  • Liquidation engine modeling and cascade analysis
  • Solvency stress testing under volatility shocks and liquidity contraction
  • Margin parameter calibration and capital efficiency optimization
  • Monte Carlo simulation of cross-market contagion and feedback dynamics
  • Incentive design and structural stability analysis
DeFi risk is inherently systemic — shaped by nonlinear feedback loops, oracle dependencies, liquidity fragmentation, and adversarial behavior. SepperLabs treats protocol risk as an engineering discipline, designing models that stress assumptions before capital is exposed to them.
Systematic Strategies & Quantitative Risk
  • Portfolio construction and factor-based allocation frameworks
  • Volatility modeling and regime-detection systems
  • Tail-risk analytics and scenario analysis
  • Statistical arbitrage and systematic signal research
  • High-dimensional simulation and performance attribution
The analytical frameworks used in DeFi stress testing are grounded in the same quantitative foundations that govern institutional portfolio risk management: correlation structure, convexity, liquidity risk, and distributional asymmetry.
Philosophy

SepperLabs emphasizes structural robustness over surface-level optimization.

The objective is not merely performance, but resilience — ensuring that strategies and protocols behave predictably across regime shifts, liquidity crises, and tail events.

Otar holds a Ph.D. in theoretical physics and applies a first-principles methodology to financial systems. Whether constructing a simulation harness for a perpetual exchange or evaluating drawdown dynamics in a systematic portfolio, the approach remains consistent:

  • Formalize the system
  • Quantify uncertainty
  • Stress the assumptions
  • Measure structural fragility
SepperLabs operates as an independent quantitative partner, delivering institutional-grade risk analytics and research tailored to both decentralized finance infrastructure and modern systematic investment strategies.
Publications & Research
Slippage-at-Risk (SaR): A Forward-Looking Liquidity Risk Framework for Perpetual Futures Exchanges
O. Sepper  ·  Preprint  ·  2026
Introduces Slippage-at-Risk (SaR), a quantitative framework for measuring liquidity risk in perpetual futures exchanges. Derives forward-looking liquidation execution risk from order book microstructure using three complementary metrics (SaR, ESaR, TSaR) with concentration adjustments for fragile liquidity structures. Includes empirical validation on Hyperliquid order book data.
arXiv preprint forthcoming
© 2026 SepperLabs