Rosheen Naeem
Software Engineer at Miro and OCF Community Member
Amsterdam, The Netherlands
Actions
I am a Software Engineer at Miro and a community member at Open Climate Fix. I completed the Erasmus Mundus Master’s in Software Engineering for the Green Deal (SE4GD), a joint degree program across Vrije Universiteit Amsterdam (Netherlands), LUT University (Finland), and Università degli Studi dell’Aquila (Italy), focused on software for environmental sustainability. I contributed to Open Climate Fix through Google Summer of Code, where I worked on transformer-based solar PV forecasting and data pipeline engineering. I’ve also contributed extensively to open source, including Google Season of Docs with the Python library ArviZ and an Outreachy internship with Open Food Facts.
Area of Expertise
Why Your Forecasting Transformer Isn’t Working (And How to Fix It in Python)
Renewable energy is clean — but it’s also inherently variable. Solar PV generation can change dramatically within minutes due to cloud cover and weather conditions, making accurate short-term forecasts essential for grid stability, energy trading, and smart-home optimisation.
Open Climate Fix builds open and high-impact forecasting tools to accelerate the transition to a low-carbon energy system. One of these projects is Open Quartz Solar Forecast: an open-source model that uses public PV generation data, site metadata, and numerical weather prediction variables to forecast solar power for any location.
In this talk, I’ll present a real case study from my Google Summer of Code project where I implemented and trained a Temporal Fusion Transformer for multi-horizon solar forecasting. I’ll cover the practical engineering challenges behind making transformer forecasting work in Python: building continuous training windows, aligning weather forecast steps with observations, separating static vs time-varying features, and stabilising training using PyTorch Forecasting and PyTorch Lightning.
Attendees will leave with reusable patterns for real-world time-series forecasting pipelines.
Quartz Solar OS: Building an open source AI solar forecast for everyone
Unlike fossil fuels, renewable energy is unpredictable in terms of power generation from one hour to the next. In order to maximise the utility of solar energy in the home and industry, being able to predict the generation coming in the next hour or day will allow smart homes to choose when to charge an EV, run a home appliance or cycle a heat pump.
OCF’s first focus is on solar short-term forecasting. Our most advanced AI model takes in real-time PV data, numeric weather predictions, satellite images, as well as regional grid data, and outputs probabilistic solar electricity nowcasts for solar production in the UK.
The talk OCF proposes will present the ML techniques used to develop Quartz Solar OS. It currently has a single-model architecture implemented and trained.
The talk will further delve into the enhancements Rosheen Naeem is working on as part of her Google Summer of Code project, which focuses on adding more model architectures. The models would be trained on publicly available weather data and a PV generation dataset that is already made publicly available.
The talk will also cover the possible use cases and climate impact such a forecast could have.
LF Energy Summit 2024 Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top