Forecast Software Engineer - Statistical Modleing & Optimization

Search Services; Houston (TX); On-site; USD 140,000 - 190,000; Full time

Job summary

Search Services is seeking a Forecast Software Engineer — Statistical Modeling & Optimization to build analytical software for ERCOT forecasting, benchmarking, and confidence scoring. You will translate established methodologies and grid-model outputs into reliable production software, blending software engineering with applied statistics and uncertainty measurement.

The role requires strong Python production skills, experience in time-series validation and backtesting, and the ability to

Qualifications

• Bachelor’s or Master’s degree in statistics, applied mathematics, operations research, computer science, engineering, econometrics, data science, or a related field.
• Strong production-level Python skills.
• Strong foundation in applied statistics and probability.
• Experience developing quantitative, analytical, or forecasting software.
• Proficiency with NumPy, Pandas, SciPy, scikit-learn, SQL, and Git.
• Experience with time-series validation, backtesting, uncertainty measurement, and model-performance analysis.
• Ability to translate domain-expert requirements into reproducible software.
• Strong testing, debugging, and technical-documentation skills.

Responsibilities

• Develop ERCOT congestion, basis, and nodal-price forecasting services.
• Implement simplified SCED, economic-dispatch, and constraint-aware optimization calculations.
• Integrate grid topology, transmission constraints, generator characteristics, load and renewable forecasts, fuel curves, and market data.
• Build rigorous historical replay and backtesting processes while preventing data leakage and look-ahead bias.
• Measure forecast accuracy, bias, stability, calibration, and uncertainty using appropriate statistical methods.
• Develop confidence-scoring methodologies based on model performance, data quality, forecast horizon, scenario volatility, model agreement, and other relevant factors.
• Build standardized frameworks for comparing internal and third-party forecasts.
• Develop ensemble-model comparisons, benchmark reports, and model-performance scorecards.
• Implement scenario and sensitivity analysis for fuel prices, load growth, renewable output, generator availability, outages, and transmission constraints.
• Develop documented APIs and batch-processing workflows for forecasting and confidence scores.
• Write maintainable, tested, version-controlled Python software and automated unit, integration, regression, and statistical tests.
• Document methodologies, formulas, assumptions, dependencies, limitations, and model versions.

Skills

Python programming

Statistics & probability

Forecasting software development

Time-series validation

SQL

Git

Education

Bachelor’s or Master’s degree in statistics / math / OR / CS / engineering

Tools

NumPy

Pandas

SciPy

scikit-learn

Pyomo

Gurobi/CPLEX/OR-Tools

PostgreSQL

Docker

REST APIs

Job description

ABOUT THE ROLE

Our Client is seeking a Forecast Software Engineer — Statistical Modeling & Optimization to build analytical software supporting ERCOT forecasting, benchmarking, and confidence scoring. Working with power-market, grid-physics, data, and software experts, this engineer will translate established methodologies and grid-model outputs into reliable production software. The role combines software engineering, applied statistics, time-series validation, optimization, and uncertainty measurement.

RESPONSIBILITIES

• Develop ERCOT congestion, basis, and nodal-price forecasting services


• Implement simplified SCED, economic-dispatch, and constraint-aware optimization calculations


• Integrate grid topology, transmission constraints, generator characteristics, load and renewable forecasts, fuel curves, and market data


• Build rigorous historical replay and backtesting processes while preventing data leakage and look-ahead bias


• Measure forecast accuracy, bias, stability, calibration, and uncertainty using appropriate statistical methods


• Develop confidence-scoring methodologies based on model performance, data quality, forecast horizon, scenario volatility, model agreement, and other relevant factors


• Build standardized frameworks for comparing internal and third-party forecasts


• Develop ensemble-model comparisons, benchmark reports, and model-performance scorecards


• Implement scenario and sensitivity analysis for fuel prices, load growth, renewable output, generator availability, outages, and transmission constraints


• Develop documented APIs and batch-processing workflows for forecasting and confidence scores


• Write maintainable, tested, version-controlled Python software and automated unit, integration, regression, and statistical tests


• Document methodologies, formulas, assumptions, dependencies, limitations, and model versions

REQUIRED QUALIFICATIONS

• Bachelor’s or master’s degree in statistics, applied mathematics, operations research, computer science, engineering, econometrics, data science, or a related field


• Strong production-level Python skills


• Strong foundation in applied statistics and probability


• Experience developing quantitative, analytical, or forecasting software


• Proficiency with NumPy, Pandas, SciPy, scikit-learn, SQL, and Git


• Experience with time-series validation, backtesting, uncertainty measurement, and model-performance analysis


• Ability to translate domain-expert requirements into reproducible software


• Strong testing, debugging, and technical-documentation skills

PREFERRED QUALIFICATIONS

• Experience with Pyomo, Gurobi, CPLEX, OR-Tools, CVXPY, or similar optimization tools


• Experience with electricity markets, utilities, grid analytics, production-cost modeling, commodity forecasting, or energy trading


• Knowledge of LMPs, congestion, economic dispatch, SCED, generator constraints, and transmission systems


• Experience with ensemble models, probabilistic forecasting, or calibrated confidence measures


• Experience with PostgreSQL, REST APIs, Docker, and cloud deployment


• Knowledge of Monte Carlo simulation, Bayesian methods, quantile forecasting, bootstrapping, or similar uncertainty-estimation techniques


• ERCOT experience is valuable but not required

WHAT SUCCESS LOOKS LIKE

• Operational ERCOT forecasting services integrated with the broader platform


• Reliable dispatch and constraint-aware forecasting capabilities


• Reproducible historical replay and backtesting


• Documented forecast-validation and confidence-scoring frameworks


• Consistent internal and third-party forecast benchmarking


• Scenario and sensitivity-analysis capabilities


• Forecast and confidence-score APIs


• Automated statistical and regression testing


• Clear documentation of methodology, assumptions, limitations, and performance

DESIRED CHARACTERISTICS

• Statistically rigorous while remaining practical and delivery-focused


• Strong software engineering discipline


• Able to identify and challenge weak validation approaches


• Comfortable explaining uncertainty and statistical results to nontechnical audiences


• Collaborative with market, data, software, and engineering specialists


• Highly attentive to reproducibility, data lineage, and model governance


• Comfortable working in a fast-moving startup environment