We are building a next-generation trading system that combines classic quantitative methods with modern AI (LLMs and agents).
I am looking for an experienced Quant / Algorithmic Trading Engineer to help design and implement the first production-grade version of this system.
This is not a “toy bot” or signal channel. The focus is on:
solid engineering,
robust risk management,
and verifiable, backtested performance.
What you’ll be building
A Python-based trading engine that can:
Connect to one or more exchanges/brokers via API (initially crypto; later futures/FX/stocks).
Ingest and store historical and live market data (candles, order books, trades).
Run rule-based and quant strategies (long/short, leverage where appropriate).
Execute orders reliably with proper logging, error handling, and safety checks.
A research / backtesting workflow, including:
Backtesting framework (Backtrader, vectorbt, Freqtrade, custom, etc.).
Walk-forward testing and out-of-sample validation.
Basic performance analytics: win rate, Sharpe, max DD, exposure, etc.
An initial strategy set, e.g.:
1–3 “production-candidate” strategies (mean reversion, breakout/trend, volatility plays, etc.).
Clear configuration and risk parameters (position sizing, per-trade loss caps, daily loss limits).
Support for both paper trading and small-size live trading.
An AI/LLM integration layer (Phase 2 of the contract):
Use LLMs/agents for:
monitoring and summarizing system health,
generating reports on strategy performance,
supporting idea generation and parameter search (human-in-the-loop).
No “GPT decides trades”; AI is an assistive layer on top of real quant logic.
Responsibilities
Work with me to refine a realistic architecture and roadmap for the system.
Implement clean, well-structured Python code for:
data ingestion and storage,
strategy execution and portfolio/risk management,
exchange/broker API connectors (REST/WebSocket).
Set up backtesting + paper trading environment and help define validation criteria.
Prototype and implement 1–3 strategies from idea → backtest → paper → small live.
Integrate LLMs/AI tools where they truly add value (e.g., using OpenAI API, LangChain, or similar) — not hype for its own sake.
Document the system so it can be extended by additional team members later.
Requirements
Please only apply if you meet most of the following:
Strong Python (data + backend):
Pandas / NumPy, async IO, REST/WebSocket APIs, testing.
Hands-on experience with algorithmic trading, ideally:
Crypto and/or FX / futures (Binance, Bybit, OKX, BitMEX, Interactive Brokers, etc.).
Practical experience with backtesting and live deployment.
Familiarity with at least one trading/backtesting framework:
Backtrader, vectorbt, Freqtrade, Zipline, QSTrader, custom, etc.
Solid understanding of risk management:
position sizing, leverage, drawdown control, kill-switches, etc.
Comfortable designing and working with a data store:
e.g., Postgres, DuckDB, or similar for storing historical data and results.
Experience integrating LLMs or ML models into applications (nice to have but not strictly required if you’re strong on quant/infra and willing to learn).
Soft stuff:
Clear communicator in English.
Comfortable collaborating over chat/voice a few times a week.
Able to work independently, propose solutions, and push things forward without micro-management.
Nice-to-have
Prior work on a crypto trading bot or prop desk tooling.
Experience with LangChain / crewAI / other agent frameworks.
Experience deploying systems on cloud/VPS (Docker, Linux).
Familiarity with event-driven architectures for trading systems. This is a hands-on engineering role. I’m not looking for a slide deck; I’m looking for working code, tested strategies, and a system we can build on.
How to apply
To help me filter out generic copy-paste proposals, please include the following in your application:
A short description (2–3 sentences) of a trading system or bot you’ve worked on:
What market(s)?
What strategy type(s)?
What was your exact role?
What stack you would choose for:
backtesting,
live execution,
data storage,
and LLM integration — and why.
One concrete example of a risk control you would implement from day one.
Proposals without these answers will likely be ignored.
If this sounds like something you’d enjoy building – and you have real experience shipping trading code, not just reading about it – I’d be happy to discuss further.
Apply Now
Apply Now