Requirements
- Have practical experience calibrating volatility surfaces on real market data
- Including handling gaps, latency issues and so on to effectively use realistic data available in the market
- Understand how to enforce smoothness, arbitrage-free conditions, and temporal stability
- Be able to tune and debug models under realistic market conditions – including bid/ask spreads, noise, and incomplete markets
- Design and implement logic for position-driven dynamic surface shaping, including:
- How current portfolio Greeks (vega, gamma, skew) should influence surface parameters such as skew, curvature, and wing behavior
- Hands-on experience is required for dynamically adapting surface shape based on current exposure
- Ability to identify, model, and mitigate residual noise in implied volatility surfaces, especially:
- near expiry,
- around illiquid strikes,
- or in event-driven conditions.
- Python (mandatory), with strong use of NumPy, pandas, matplotlib, SciPy, and relevant optimization/ML libraries
- Familiarity with standard quant libraries (QuantLib, or custom volatility tools)
Nice to Have
- MFT’ish research is must.
- HFT is nice to have.
- PyTorch / TensorFlow experience (strongly preferred)
- Experience with NSE options and/or other TradFi derivatives with margin impact is a major plus.
- Familiarity with practical heuristics for surface management
- Working (not just academic) experience applying ML/DL models (e.g., PyTorch, TensorFlow) to this problem
- Understanding of model explainability and risk of overfiting in execution-sensitive environments
- Direct experience in spot/futures vs. options arbitrage
