Crypto Trading Bot: Strategy Developer (Python/React)
Budget / Salary$1,500–3,000
TypeFreelance project
LocationRemote
Posted1 hour ago
Quant Developer for a Crypto Trading Bot: Strategy Research and Development (Python, React/Node)
About the project
A working crypto trading bot with a web dashboard. It paper-trades 12 coins on exchange data (15m to 1D candles) and can trade real funds on Base through Uniswap, with every trade confirmed in MetaMask. It already has:
12 strategies (RSI, Bollinger, Supertrend, trend filter, momentum, whale-flow copy trading and more)
backtesting with fees and slippage
an optimizer that tunes on older data and tests on newer data it never saw
Auto-pick, which chooses the strategy that held up on that unseen data
What you'll do
Research, build and test new trading strategies, and improve the existing ones.
Prove each strategy holds up on data it wasn't tuned on (walk-forward testing, robustness across settings, realistic fees) before it goes live.
Improve risk management: position sizing, stops, exposure limits and drawdown control.
Extend the analysis: performance metrics, reports and on-chain whale-flow analysis.
Keep the code clean and tested (the project has automated checks for Python and the frontend).
Requirements
A proven track record designing and backtesting trading strategies, with examples you can show.
A deep understanding of overfitting, look-ahead bias, fees, slippage and the difference between results on past data and on new data.
Strong Python (no pandas in the core today) and solid JavaScript/Node (React and Vite).
Data analysis skills: statistics, time series and judging whether a result is significant.
Clear written communication in English.
Nice to have
Freqtrade or similar bot frameworks.
DeFi and EVM experience (Uniswap v3, viem, on-chain data).
Crypto market structure and order-flow analysis.
To apply, send
A strategy you built and how you validated it, with results on data it wasn't tuned on.
How you'd decide whether a new strategy is ready to trade real money.
Your availability and rate.
Note: we're looking for sound, honestly tested strategies, not promises of guaranteed profit.
About the project
A working crypto trading bot with a web dashboard. It paper-trades 12 coins on exchange data (15m to 1D candles) and can trade real funds on Base through Uniswap, with every trade confirmed in MetaMask. It already has:
12 strategies (RSI, Bollinger, Supertrend, trend filter, momentum, whale-flow copy trading and more)
backtesting with fees and slippage
an optimizer that tunes on older data and tests on newer data it never saw
Auto-pick, which chooses the strategy that held up on that unseen data
What you'll do
Research, build and test new trading strategies, and improve the existing ones.
Prove each strategy holds up on data it wasn't tuned on (walk-forward testing, robustness across settings, realistic fees) before it goes live.
Improve risk management: position sizing, stops, exposure limits and drawdown control.
Extend the analysis: performance metrics, reports and on-chain whale-flow analysis.
Keep the code clean and tested (the project has automated checks for Python and the frontend).
Requirements
A proven track record designing and backtesting trading strategies, with examples you can show.
A deep understanding of overfitting, look-ahead bias, fees, slippage and the difference between results on past data and on new data.
Strong Python (no pandas in the core today) and solid JavaScript/Node (React and Vite).
Data analysis skills: statistics, time series and judging whether a result is significant.
Clear written communication in English.
Nice to have
Freqtrade or similar bot frameworks.
DeFi and EVM experience (Uniswap v3, viem, on-chain data).
Crypto market structure and order-flow analysis.
To apply, send
A strategy you built and how you validated it, with results on data it wasn't tuned on.
How you'd decide whether a new strategy is ready to trade real money.
Your availability and rate.
Note: we're looking for sound, honestly tested strategies, not promises of guaranteed profit.
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.