PhD candidate at University College London (UCL) focusing on:

• Quantitative energy systems
• AI / ML for physical assets
• ESG- and climate-aligned investment modelling

I sit at the intersection of investment and technology:

• Build models and pipelines that link physical energy assets (batteries, PEM electrolysers, hybrid storage)
  with financial outcomes (cash flows, risk, ESG metrics, scenario analysis).
• Work with simulation, optimisation, and machine learning to support decisions in
  energy trading, infra investing, and portfolio risk management.

• Investment & Risk

  • Scenario-based valuation of energy assets (batteries, hydrogen, hybrid storage)
  • ESG risk analytics, carbon accounting, EU Taxonomy & SFDR alignment
  • Portfolio stress testing, VaR, Sharpe ratio, factor and sensitivity analysis

• Technology & Modelling

  • Multiscale electrochemical modelling (PEMWE, Li-ion, HESS)
  • Physics-based + data-driven hybrid models (COMSOL, PyBaMM, Simulink)
  • Scientific ML: PINNs, surrogate models, GAN-based augmentation

• Systems & Infrastructure

  • Data centre energy resilience and hybrid storage sizing
  • Power system modelling (grid interaction, ancillary services)
  • Long-duration storage and hydrogen system integration

Programming & Data

• Python (NumPy, Pandas, Scikit-learn, XGBoost, Matplotlib, Plotly)
• MATLAB / Simulink / Simscape
• SQL, Bash
• Basic Julia & C++

Modelling & ML

• Time-series modelling, regression, feature engineering
• Scenario generation, Monte Carlo, sensitivity analysis
• GANs for data augmentation, surrogate modelling for simulations
• PyBaMM (battery models), COMSOL (P2D/DFN), Simulink (system-level)

Finance & ESG

• Portfolio analytics: VaR, CVaR, stress testing, performance attribution
• ESG reporting: EU Taxonomy, SFDR-aligned metrics, carbon cost modelling
• Cash-flow modelling for energy projects and structured products
• Understanding of energy markets (TOU arbitrage, ancillary services, power pricing)

Energy & Engineering

• PEM water electrolysers, Li-ion batteries, hybrid energy storage systems (HESS)
• Data centre load modelling (mission-critical infrastructure)
• GITT, EIS, OCV analysis, electrochemical parameter identification

• PhD, Quantitative Energy Systems — University College London (UCL)
  Multiscale modelling & optimisation of hybrid electrochemical energy storage systems for data centre resilience.

• M.Res., Control Engineering — University of Warwick
  Quantitative modelling & system optimisation.

• M.Sc., Electrical & Electronic Engineering — University of Nottingham
  Data analytics & simulation.

• B.Eng., Energy & Power Engineering — Southeast University
  Outstanding Bachelor Graduate.

Nuremberg, Germany · 2024

• Built scenario-based valuation models for structured products linked to battery and carbon assets.
• Integrated pricing data, ESG risk factors, lifecycle value-at-risk into model outputs.
• Automated Python / MATLAB / Excel VBA pipelines for:
  • portfolio stress testing
  • performance attribution
  • compliance checks for ESG-focused multi-asset funds.
• Collaborated with portfolio managers and risk teams to:
  • quantify time-of-use arbitrage
  • estimate residual value of battery assets
  • design hedging strategies aligned with ESG mandates.
• Produced internal reports tied to EU Taxonomy and SFDR, enabling integration with trading desks and compliance.

Nuremberg, Germany · 2024

• Developed physics-based COMSOL + PyBaMM models for fast-charging strategies under ESG constraints
  (thermal impact, energy efficiency, lifecycle impact).
• Performed scenario-based stress testing and sensitivity analysis to estimate:
  • marginal carbon cost
  • time-of-use arbitrage potential
  • residual value impact across charging strategies.
• Automated simulation pipelines for ESG-linked structured products and smart BMS strategies,
  enabling integration with carbon trading platforms.

Munich, Germany & Birmingham, UK · 2021–2023

• Constructed dynamic dispatch and cash-flow models for hydrogen and power assets under:
  • variable market pricing
  • ancillary service provision
  • multi-asset allocation logics.
• Conducted sensitivity analysis on 10k+ datapoints/run:
  • extracted efficiency, degradation, and return profiles
  • benchmarked opportunities under EU Fit-for-55 scenarios.
• Built Python-based analytics to:
  • quantify marginal carbon cost
  • simulate ESG-aligned investment scenarios
  • support allocation across green infrastructure portfolios.
• Contributed to ESG-compliant capital deployment into hydrogen and power projects.

• XPeng Europe — Business analysis on energy & product systems
• WMG, University of Warwick — AI simulation research assistant
  Multiscale model + ML optimisation for electrochemical systems

Python · SQL · ML · Simulink

• Built multi-factor models simulating portfolio exposure to:
  • interest rates
  • FX
  • equities
  • commodities.
• Calibrated on 150+ market data sources (>10k datapoints), achieving <3% forecast error for return projections and VaR.
• Designed an optimisation engine improving risk-adjusted performance (Sharpe +12.5%) in ESG-aligned portfolios.
• Automated cash-flow reconciliation and reporting, reducing manual workload by ~70%.

Simulink · Simscape · Python · GANs

• Developed integrated Simulink/Simscape models for PEM electrolysers under:
  • 1.8 A/cm²
  • 80°C
  • 30 bar.
• Achieved <3% simulation error vs experimental data; voltage efficiency 76% @ 1.5 A/cm², thermal efficiency 63.4%.
• Extracted parameters from 150+ studies for data-driven calibration.
• Used GAN-based augmentation to expand dataset from 822 → 12,300 samples, training ML models with:
  • R² = 0.94
  • MAE = 0.028 V.
• Outputs feed into asset valuation and resilience studies for hybrid systems and data centres.

MATLAB/Simulink · Python · Optimisation

• Designed and simulated a PEM-integrated hybrid energy storage system (HESS) for data centres.
• Evaluated trade-offs between:
  • energy capacity
  • charge/discharge power
  • unmet load
  • grid imports.
• Identified an approximate Pareto knee at:
  • 2,000 kWh storage
  • 500 kW discharge power
    where resilience gains flatten and marginal benefit of extra investment is negligible.
• Framework supports sizing decisions for infra and data-centre-focused investors.

Building investment-grade models for real-world energy systems,
where physical realism, regulatory alignment, and financial performance are jointly optimised.