Prediction Market Quantitative Engineer

G-20 Group•New York, NY

11h

About The Position

About G20 Group The G-20 Group is a pioneer in Quantitative Trading systems in cross-asset markets. Headquartered in Switzerland, we operate at the intersection of Quantitative Research, Software Engineering and Trading. The team combines a startup mindset with extensive experience in proprietary Trading, Technology and Quantitative Finance. Role Overview We are hiring a Prediction Market Quant Engineer to build research and trading infrastructure for operating in prediction markets (event contracts) across multiple venues. You will design models that estimate event probabilities, detect mispricing, size positions, and manage risk – then translate them into reliable systems that run end-to-end (data → forecasting → execution → monitoring). This role sits at the intersection of quant research, engineering, and market microstructure, and is ideal for someone who enjoys shipping robust systems as much as developing models.

Requirements

Degree in Quantitative Finance, Mathematics, Computer Science, Statistics, or a related quantitative field.
Strong engineering skills with Python (required); experience with production systems and data engineering.
Solid foundation in statistics, probability, and machine learning (calibration, uncertainty, causal pitfalls, time-series).
Experience building backtests and evaluating predictive models with appropriate metrics (e.g., log loss/Brier, calibration).
Familiarity with trading concepts: expected value, position sizing, risk budgeting, correlation, liquidity constraints.
Ability to communicate clearly about model assumptions, limitations, and risk.
Some schedule flexibility may be required around major event windows
Self-motivated, detail-oriented, and comfortable working in a dynamic, startup-like environment.
Only candidates who possess the pre-existing right to work in one of the locations above without company sponsorship need apply.

Nice To Haves

Prior work in forecasting, sports analytics, political modeling, event-driven trading, or market-making/liquidity modeling.
Experience with NLP for news/social/media signals; knowledge graphs or information retrieval for event resolution.
Knowledge of prediction market mechanics (order books vs AMMs, fee structures, market manipulation/anti-manipulation signals).
Proficiency with SQL; experience with streaming systems (Kafka), workflow orchestration (Airflow), and cloud (AWS/GCP/Azure).
Experience with Bayesian methods, probabilistic programming (Stan/PyMC), or ensemble methods.
Familiarity with rigorous experimentation: online/offline evaluation, data leakage prevention, and model governance.

Responsibilities

Modeling & Research Develop probabilistic models to forecast outcomes of real-world events (e.g., elections, macro releases, sports, policy decisions, industry milestones).
Combine heterogeneous signals (time series, text/news, market data, polling/alternative data, fundamentals, expert priors) into calibrated probability estimates.
Build pricing and edge frameworks: fair value, uncertainty bands, expected value, and model drift/regime diagnostics.
Design evaluation methods (proper scoring rules like log loss/Brier score, calibration curves, back-tests with realistic costs and constraints).
Trading & Market Design (Applied) Identify and exploit mis-pricings across contracts/venues; design cross-market arbitrage and relative-value strategies where feasible.
Build position sizing and risk frameworks (Kelly variants, drawdown/risk budgets, scenario stress tests, liquidity/impact-aware sizing).
For multi-outcome markets: enforce probability coherence (no-arb constraints, normalization) and portfolio optimization across correlated contracts.
Engineering & Production Build data pipelines and real-time services for ingesting, cleaning, and versioning market + external data.
Implement execution tooling: order management, smart routing (where applicable), monitoring, and automated safeguards.
Create dashboards/alerts for performance, exposure, model health (calibration, drift), and operational integrity.
Ensure reproducibility: experiment tracking, model registry, CI/CD, and robust testing.
Collaboration & Governance Work closely with trading/risk/compliance stakeholders to translate research into controlled deployment.
Document models, assumptions, failure modes, and operating procedures; participate in incident reviews and continuous improvement.