Kubu Labs

Non-trivial systems. Real finance.

Applied research in portfolio optimization with a focus on Web3 asset allocation. Quant models, on-chain risk and practical execution in production environments.

View repository
✔ On-chain risk metrics⚡ Low-latency execution

Research

Multi-factor portfolios
Momentum, carry, relative value and liquidity for on-chain universes (L1s, L2s, DeFi blue chips).
Web3 risk management
Protocol, smart-contract, oracle and MEV risks; base-volatility and adaptive position sizing.
Execution & costs
Slippage, gas, swap routes, cross-chain liquidity, rebalance windows and market impact.

Methodology

  • Modeling: Mean-variance (Markowitz), Risk Parity, CVaR, fractional Kelly.
  • Data: On/off-chain prices, TVL, DEX/CEX liquidity, funding, fees, protocol risks.
  • Backtesting: Rolling windows, walk-forward, regime drift, realistic frictions and costs.
  • Execution: DEX aggregators/RFQ, multi-pool routes, controlled slippage, programmatic rebalance.

Stack

  • Data: Subgraphs/ETL, ClickHouse/Parquet, time-series store.
  • Quant: Python (NumPy/Pandas), Rust (critical paths), JAX/PyTorch (optimization).
  • On-chain: RETH/Foundry/ethers-rs; oracles and monitoring.
  • App: Static HTML + later API, lightweight dashboards (SVG/Canvas).

Get the monthly technical brief

Quant insights, execution lessons and risk metrics — straight to your inbox.

FAQ

Do you manage capital or only publish research?
Focus on applied research and tooling. This is not investment advice.
Which assets are in scope?
Majors (BTC/ETH), selected L1s/L2s and DeFi blue chips with adequate liquidity.
How do you account for costs and risks?
Transaction costs, gas, slippage, protocol and execution risks are modeled.
Can I integrate my own data?
Yes: connect your warehouse/lake and run your own backtests.