I will develop quantitative trading systems for polymarket and kalshi
Quantitative Developer: Research, Backtest and Deploy Trading Systems
Level 2
Has met high performance criteria and has a proven track record for meeting client expectations.
About this Gig
I turn prediction market theses and inefficiencies into proprietary quantitative systems for Polymarket and Kalshi, combining statistical rigor, robust engineering, and operational efficiency.
I work across the full lifecycle, from hypothesis formulation and data acquisition to modeling, validation, automation, and deployment. A project can begin with a thesis, an existing strategy, or a market opportunity.
Depending on the scope, I develop data pipelines, feature engineering, probabilistic models, event-driven research, order book analysis, REST/WebSocket integrations, realistic backtesting, out-of-sample testing, robustness analysis, risk controls, execution logic, monitoring, and logging.
Implementation is primarily in Python, with modular architecture, clean code, auditable assumptions, and technical documentation.
I do not offer generic bots or guarantee profits. My work turns hypotheses into reliable, testable quantitative infrastructure built for professional prediction market operations.
NDA available. Before ordering, send your hypothesis, target markets, data, execution requirements, infrastructure, and project objective.
Programming language:
Python
•
SQL
Frameworks:
Scikit-learn
•
PyTorch
•
Panda
•
Other
APIs:
Other
Tools:
Jupyter Notebook
•
TensorFlow
FAQ
What exactly can be included in the project?
Depending on the package, the project may include quantitative research, data, probabilistic modeling, Python engineering, signals, backtesting, validation, risk controls, APIs, automated execution, logging, monitoring, documentation, and deployment assistance.
What is the difference between the packages?
The Blueprint defines feasibility, hypothesis, model, data, risk, architecture, and technical specifications. The Prototype delivers a Python system with signals, backtesting, and validation. The Production package adds APIs, automated execution, logging, monitoring, and deployment assistance.
Do you cover the entire quantitative development lifecycle?
Yes. I cover the entire lifecycle, from hypothesis formulation and data structuring to modeling, validation, engineering, execution, and deployment. Technical depth and operational readiness vary by package.
Do you work exclusively with Polymarket and Kalshi?
Polymarket and Kalshi are my primary focus. I can also develop systems for other prediction markets when they provide APIs, data, and infrastructure compatible with the project’s technical and operational requirements.
What technology stack do you use?
I primarily use Python, SQL, pandas, NumPy, statsmodels, and scikit-learn. When needed, I also work with XGBoost, PyTorch, REST APIs, WebSockets, CLOBs/order books, databases, Web3 tools, logging, and monitoring.
How do you validate a strategy?
Validation may include realistic backtesting, in-sample/out-of-sample testing, walk-forward analysis, Monte Carlo simulations, sensitivity analysis, stress testing, costs, slippage, liquidity, probabilistic calibration, Brier score/log loss, stability, and assumption audits.
Are the systems truly production-ready?
Yes, with the top-tier package. Deliverables may include APIs, automated execution, state management, error handling, risk controls, logs, alerts, monitoring, documentation, and deployment assistance. Live operation depends on the client’s infrastructure and credentials.
What do I need to provide to get started?
Send your hypothesis or strategy, target markets and events, data sources, execution requirements, infrastructure, operational constraints, and end goal. If starting from scratch, I can also help structure the hypothesis and research protocol.
Do you guarantee profits, ROI, or future performance?
No. I do not promise profits, ROI, win rates, or future performance. My work is driven by evidence, statistical rigor, sound engineering, and risk management. Historical, simulated, or validated results do not guarantee live performance.
How do you handle communication and confidentiality?
Communication occurs primarily through Fiverr chat and written documentation, ensuring accuracy, traceability, and auditable decisions. Short calls may be used for alignment or demonstrations. I can work under an NDA, and your strategy, code, data, documentation, and results remain confidential.

