QUANTITATIVE DEFI PRODUCT

PRODUCT + RESEARCH

Product Manager & Quantitative Analyst

One Click
Labs

Applying portfolio thinking to fragmented DeFi yield markets.

A quantitative product combining API-driven market data, risk analysis and portfolio optimisation to help users evaluate yield opportunities as part of a portfolio rather than as isolated APY figures.

  • Product management
  • API integrations
  • DeFiLlama
  • Python
  • Portfolio optimisation
  • Risk modelling
01

THE PROBLEM

Thousands of yields. No common decision framework.

Lending pools, staking positions, liquidity pools and automated vaults existed across many protocols and blockchains, but the information required to compare them was fragmented.

APYs changed rapidly, pool data used inconsistent formats and historical performance was incomplete. Similar headline yields could conceal very different return mechanics and risk exposures.

A 20% lending yield, a 20% liquidity-pool yield and a 20% incentive-driven vault were not equivalent investments.
  • Protocol and smart-contract risk
  • Asset volatility
  • Liquidity and withdrawal constraints
  • Yield instability and incentive dependence
  • Impermanent loss
  • Concentration
  • Blockchain and bridge exposure

The highest advertised yield was rarely the best portfolio decision.

02

WHAT I WORKED ON

From raw market data to portfolio decisions

The work connected live-market infrastructure, quantitative research, product definition and practical strategy management.

01

Data and API infrastructure

  • DeFiLlama and protocol APIs
  • Python and SQL workflows
  • Extraction, cleaning and normalisation
  • Historical APY and TVL data
  • Validation and exception handling
02

Quantitative research

  • Expected return estimation
  • Volatility and downside analysis
  • Strategy correlation
  • Risk scoring
  • Efficient-frontier construction
  • Backtesting and portfolio comparison
03

Product development

  • User and product requirements
  • Portfolio-building workflow
  • Quantitative outputs translated into features
  • Data, research and interface prioritisation
  • Product and technical coordination
04

Live strategy management

  • Active yield-strategy monitoring
  • Yield and risk reassessment
  • Market feedback applied to allocation logic
  • Model outputs connected with practical decisions

My role sat between the data, the model and the user-facing product.

03

HOW THE SYSTEM WORKED

A comparable layer over inconsistent markets

Protocol-level information had to be extracted, validated and transformed before it could support risk-aware portfolio decisions.

VISUAL 01Data-to-portfolio architectureSystem schematic
  1. 01DeFi protocols + DeFiLlama
  2. 02API ingestion
  3. 03Cleaning and normalisation
  4. 04Yield + risk features
  5. 05Portfolio optimisation
  6. 06Backtest and review
  7. 07User-facing portfolio

Quantitative outputs informed decisions; they did not replace protocol diligence or judgement.

04

VISUALISING PORTFOLIO OPTIMISATION

Risk and return behaved differently across strategies

The models helped structure comparisons while remaining dependent on estimates, changing markets and protocol-level diligence.

PORTFOLIO CONSTRUCTION

More efficient combinations, not maximum APY

The frontier represents the portfolios with the highest estimated return for each level of risk. Inputs remained estimates, not promises.

VISUAL 02Illustrative portfolio optimisationNo historical performance shown
Illustrative efficient frontier chartEstimated risk increases from left to right and expected portfolio return increases from bottom to top. Yield strategies appear as scattered opportunities, with an efficient frontier and one selected risk-adjusted portfolio.LendingStakingVaultLP strategyIncentivised poolSelected portfolioESTIMATED RISK →EXPECTED PORTFOLIO RETURN →EFFICIENT FRONTIERINEFFICIENT COMBINATIONS

The frontier represents portfolios offering the highest estimated return for a given level of risk. The objective was to identify more efficient combinations—not simply maximise APY.

RETURN MECHANICS

Liquidity provision behaved differently from lending

Displayed fee yield was only one part of the outcome. Relative price movement continuously changed the pool composition.

VISUAL 03Illustrative liquidity-provider mechanicsStandard constant-product relationship
Impermanent loss by asset-price ratioImpermanent loss is zero at the balanced entry ratio of one and becomes more negative as the relative asset price moves in either direction.BALANCED ENTRY / 1.0ASSET-PRICE RATIO RELATIVE TO ENTRY →IMPERMANENT LOSS VS HOLDING →0%−10%−20%0.251.04.0

Liquidity-provision yield had to be evaluated alongside price divergence, pool rebalancing and impermanent loss—not treated like conventional interest income. Fee yield could offset part of the loss, but did not remove the underlying exposure.

05

FROM RESEARCH PRODUCT TO OPERATING BUSINESS

A market problem that continued to evolve

The initial work focused on helping users discover and optimise DeFi yield portfolios. As the market developed, the company moved further toward yield distribution, liquidity formation and connecting protocols with allocators.

01

One Click Labs

Yield discovery and portfolio optimisation

02

Market evolution

Greater demand for structured distribution and liquidity

03

Yield Network

Active yield and on-chain liquidity platform

The original product and market research contributed to a wider company evolution toward what now operates as Yield Network.

The current business is a continuation of the company’s broader evolution, not a product I claim to have built or currently operate.

Project continuationView Yield Network ↗
06

WHAT THE PROJECT DEMONSTRATES

Quantitative product work in a live emerging market

01

Quantitative product thinking

Turning financial theory into product logic and user decisions.

02

Data systems

Integrating, cleaning and structuring fragmented API data from live markets.

03

Financial modelling

Applying risk, return, correlation and portfolio optimisation to new asset structures.

04

Cross-functional execution

Working between research, product, engineering and live investment activity.

One Click Labs was where I first combined product ownership, market-data systems and quantitative finance inside a live emerging-market product.