Francisco is available for hire
Hire FranciscoFrancisco Javier Ferrada Ferrada
Verified Expert in Engineering
Operations Research and Quantitative Analytics Specialist and Developer
Santiago, Chile
Toptal member since July 10, 2026
Bio
Francisco is an operations research and quantitative analytics specialist with 4+ years of experience in optimization and forecasting for Manakin Energy, Evalueserve, and Pacific Hydro Chile. His primary expertise is in SQL, Python, and Excel for fintech, energy, and technology startups, where he thrives in fast-paced environments. Francisco built a live SaaS platform for real-time energy-market intelligence while at Manakin Energy.
Portfolio
Manakin Energy
Machine Learning Operations (MLOps), LightGBM, Dynamic Programming, ETL, Pandas...
Evalueserve
SQL, Python, Databricks, Pandas, Industrial Engineering, Models, Git...
Pacific Hydro Chile
Python, Models, Linear Optimization, Unit Commitment Modeling, Pandas...
Experience
- Data Analytics - 6 years
- Data Analysis - 6 years
- Pandas - 5 years
- Python - 5 years
- Machine Learning - 4 years
- SQL - 4 years
- Linear Programming - 4 years
- Machine Learning Operations (MLOps) - 3 years
Preferred Environment
Databricks, Docker, Python, SQL
The most amazing...
...SaaS platform I've built is Manakin Energy, which delivers real-time energy market intelligence to battery storage operators and power generators.
Work Experience
Founder and Lead Engineer
2026 - PRESENT
Manakin Energy
- Founded and built a production SaaS platform for real-time energy-market intelligence targeting BESS (battery storage) operators and power generators in Chile's National Electric System (SEN).
- Engineered an ML price-forecasting pipeline using LightGBM to produce predictions across 50 time horizons, from 15 minutes to 24 hours, integrating live data sources like PLEXOS PDO/PID, RIO, CMG, and TCO.
- Implemented battery-dispatch optimization using dynamic programming, generating revenue-maximizing operating policies for storage assets.
- Owned the full product lifecycle end to end and automated ETL pipelines, analytics, and a multi-tenant web product in a fast-paced startup setting.
Technologies: Machine Learning Operations (MLOps), LightGBM, Dynamic Programming, ETL, Pandas, Machine Learning, Industrial Engineering, FastAPI, Models, Git, PostgreSQL, Google Sheets, Data Visualization, Artificial Intelligence (AI), Predictive Modeling, Data Analytics, Data Analysis, Analysis, Data Science, Analytics, Microsoft Access, Data Engineering, Data Migration, Data Synchronization, Technical Documentation, Data Modeling
Quantitative Developer, Risk and Quant Solutions
2025 - PRESENT
Evalueserve
- Re-engineered legacy SQL and data pipelines into scalable Python workflows on Databricks, cutting execution time and improving the reliability of reporting.
- Automated complex analytical pipelines, reducing manual processing hours and increasing data accuracy, resulting in direct operational efficiency gains. Monitored data quality and performance metrics.
- Partnered with cross-functional, distributed teams to deliver analyses on time and communicated results to non-technical stakeholders.
Technologies: SQL, Python, Databricks, Pandas, Industrial Engineering, Models, Git, PostgreSQL, Spark, Snowflake, Google Sheets, Data Visualization, Predictive Modeling, Data Analytics, Data Analysis, Analysis, Analytics, Data Engineering, Data Migration, Data Synchronization, SQL Server, Technical Documentation, Data Modeling
Energy Market and Quantitative Analyst
2023 - 2025
Pacific Hydro Chile
- Built and automated dispatch-optimization models for wind-farm battery storage using predictive statistical models, improving operational revenue.
- Led short-term electricity price-forecasting projects and served as a technical stakeholder for forecast deliverables.
- Developed in-house portfolio-optimization and unit-commitment models for scenario and sensitivity analysis of the national electricity system, supporting investment decisions.
- Produced engineering management reports on the impact of market conditions on generation assets and PPAs.
Technologies: Python, Models, Linear Optimization, Unit Commitment Modeling, Pandas, Linear Programming, Machine Learning, Industrial Engineering, SciPy, Git, Google Sheets, Data Visualization, Artificial Intelligence (AI), Predictive Modeling, Data Analytics, Data Analysis, Analysis, Data Science, Analytics, Microsoft Access, Data Migration, Data Synchronization, Technical Documentation, Data Modeling
Quantitative Portfolio and Risk Engineer
2022 - 2023
Xepelin
- Designed and implemented credit-line allocation models for SME lending operations in Mexico, improving portfolio risk-adjusted returns.
- Automated 80% of risk operations through systematic implementation of key risk variables, reducing manual workload and decision latency.
- Developed VaR, CVaR, and Monte Carlo simulations for default-probability forecasting across a dynamic loan portfolio.
- Built data-driven parameter-monitoring systems supporting real-time risk assessment and credit-policy decisions.
Technologies: Python, SQL, Google BigQuery, Looker Studio, Machine Learning, Pandas, Linear Programming, Industrial Engineering, Google Cloud Platform (GCP), Models, SciPy, Git, Google Sheets, Data Visualization, Data Analytics, Data Analysis, Analysis, Data Science, Analytics, Data Engineering, Data Migration, Data Synchronization, Technical Documentation, Data Modeling
Operations Research Project Engineer
2021 - 2023
ANILLO Project, ANID | University of Adolfo Ibáñez
- Decomposed ambiguous, large-scale planning problems into tractable optimization models spanning Chile's electric system.
- Co-authored 3 peer-reviewed papers translating complex technical findings for diverse academic and industry audiences.
- Collaborated with a multidisciplinary research team to structure long-term energy-planning studies used in academic publications.
Technologies: Python, AMPL, Mixed-integer Linear Programming, Linear Programming, Pandas, Machine Learning, Industrial Engineering, Unit Commitment Modeling, Models, Google Sheets, Data Visualization, Data Analytics, Data Analysis, Analysis, Analytics
Experience
BESS | Battery Energy Storage Dispatch Optimizer
https://github.com/fjferrada/battery-dispatch-optimizerBuilt this project independently to demonstrate the optimization techniques I apply professionally in energy markets, using synthetic data.
The project solves the battery energy storage (BESS) dispatch problem: given an hourly price forecast, it determines the optimal charge, discharge, and hold schedule to maximize arbitrage revenue over a time horizon. It uses exact dynamic programming with backward induction over a discretized state-of-charge grid, followed by a forward pass to reconstruct the optimal operating schedule. The model incorporates realistic operational constraints, including state-of-charge limits, maximum charge/discharge power, round-trip efficiency losses, and a cyclical constraint requiring the battery to return to its initial state of charge by the end of the horizon.
This approach provides a lightweight way to solve the optimization problem without requiring a solver. The main trade-off is discretizing a problem that could otherwise be continuous.
TECHNIQUES AND TOOLS
Dynamic programming, discrete-state optimization, time-series price modeling, and Python (NumPy, Pandas, and Matplotlib)
The project solves the battery energy storage (BESS) dispatch problem: given an hourly price forecast, it determines the optimal charge, discharge, and hold schedule to maximize arbitrage revenue over a time horizon. It uses exact dynamic programming with backward induction over a discretized state-of-charge grid, followed by a forward pass to reconstruct the optimal operating schedule. The model incorporates realistic operational constraints, including state-of-charge limits, maximum charge/discharge power, round-trip efficiency losses, and a cyclical constraint requiring the battery to return to its initial state of charge by the end of the horizon.
This approach provides a lightweight way to solve the optimization problem without requiring a solver. The main trade-off is discretizing a problem that could otherwise be continuous.
TECHNIQUES AND TOOLS
Dynamic programming, discrete-state optimization, time-series price modeling, and Python (NumPy, Pandas, and Matplotlib)
Risk Engine
https://github.com/fjferrada/risk-engineHistorical value-at-risk (VaR) and portfolio risk analytics for a multi-asset portfolio. Recently, I added an agent feature to ask about the results of risk metrics.
RiskEngine lets you build a portfolio (dollars invested per asset) and instantly get a professional-grade risk report: VaR by three methodologies, Conditional VaR (Expected Shortfall), a full suite of risk/return metrics, per-asset risk attribution, a correlation matrix, and interactive charts. Market data is ingested from Yahoo Finance via yfinance.
RiskEngine lets you build a portfolio (dollars invested per asset) and instantly get a professional-grade risk report: VaR by three methodologies, Conditional VaR (Expected Shortfall), a full suite of risk/return metrics, per-asset risk attribution, a correlation matrix, and interactive charts. Market data is ingested from Yahoo Finance via yfinance.
Manakin Energy
https://manakinenergy.comA production SaaS platform delivering real-time energy-market intelligence to battery-storage operators and generators in Chile. ML price forecasting across 50 horizons plus dynamic-programming dispatch optimization, all in a multi-tenant full-stack system.
Education
2020 - 2022
Master's Degree in Industrial Engineering and Operations Research
University of Adolfo Ibáñez - Santiago, Chile
2016 - 2020
Bachelor's Degree in Industrial Engineering
University of Adolfo Ibáñez - Santiago, Chile
Skills
Libraries/APIs
Pandas, NumPy, Scikit-learn, SciPy, Matplotlib
Tools
Microsoft Access, Google Sheets, Git, Pytest, Plotly
Languages
Python, SQL, Snowflake, Julia, AMPL
Paradigms
Dynamic Programming, Linear Programming, ETL
Frameworks
LightGBM, Spark
Storage
Data Synchronization, PostgreSQL
Platforms
Databricks, Docker, Google Cloud Platform (GCP)
Other
Data Visualization, Predictive Modeling, Data Analytics, Data Analysis, Analysis, Analytics, Unit Commitment Modeling, Machine Learning, Artificial Intelligence (AI), Data Science, Data Engineering, Data Migration, SQL Server, Technical Documentation, Data Modeling, Machine Learning Operations (MLOps), VaR, FastAPI, Models, Linear Optimization, Mixed-integer Linear Programming, Statistics, Operations Research, Industrial Engineering, Google BigQuery, Looker Studio, Financial Risk Management, Feature Engineering, LLM Agents, AI Agents
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring