Google Summer of Code 2025: Forecasting Global Solar Sites Using TZ-SAM Data - Discussion Thread

[zakwatts]

Welcome and thanks for taking an interest in this GSOC 2025 project. This discussion thread is here for you to ask any questions you have about this project. I'm happy to provide clarifications and help you understand the project's goals, scope, and requirements. Also please feel free to ask about anything that interests you!

Solar Forecast with TZ-SAM

Create Solar Forecasts for a Site or a Grid/Country Based on Transition Zero’s Solar Asset Mapper Data

Project Description

The Transition Zero Solar Asset Mapper (TZ-SAM) is a new open source tool launched by not-for-profit Transition Zero (TZ). It provides a database of large solar farms, around 60,00 in total, from around the world that have been identified by using advanced AI satellite imagery identification. We want to potentially use the TZ Solar asset mapper dataset to forecast each of the sites it identifies and/or map out total capacities for large sites each country.

Possible Outcomes

Here are some ideas for potential outcomes. I am very open to new and innovative ideas on ways to use this dataset in regards to solar forecasting so please feel free to reach out with any ideas you may have! One possible outcome is to create a basic solar forecasts, using Open Source Quartz Solar Forecast for every site in the TZ-SAM dataset using a global NWP provider. Some analysis of the TZ-SAM dataset could be useful for this. This could then run frequently and have a basic UI to show forecasts for each of the sites. Evaluating the model performance on real world data would also be a useful test. It would especially be interested to see how errors in solar forecasts vary across different regions.

Other Key Information

Expected Size: 175hrs
Skills: Running ML models, docker (not essential but could be useful)
Difficulty level: Medium
Related Reading: https://www.transitionzero.org/products/solar-asset-mapper, https://github.com/openclimatefix/open-source-quartz-solar-forecast
Potential mentors: @zakwatts

Please note that this discussion is for questions and clarifications, not for formal applications.

[itsbharatj]

Hey @zakwatts, I find this project very exciting and would like to learn more about it. I would like myself to get familialized by https://github.com/openclimatefix/open-source-quartz-solar-forecast codebase. Would you recommend some tasks which we could do the codebase to better undersatnd it.

[zakwatts]

Hi @itsbharatj, thanks for you interest in this project. Familiarising yourself with Open Quartz repo you linked if a good place to start.

I'd defiantly have a got at running at running the forecat, perhaps with the different models. One possible route to this model could be to adapt it be predict for countries. I think getting familiar with the 'run forecast' functions would be a good place to start.

[itsbharatj]

Sounds great; I would be running the forecast model for different countries. Should give me a good familiarity with the project and codebase.
Do you have an internal slack or discord for me to better engage with the community?

[zakwatts]

For the moment we want to keep most of the discussions on Github so they are available for all to see :)

[PredictiveManish]

Hey @zakwatts, this project is pretty good and I am interested in working and contributing to this project, I will be very grateful if you can share something which is must to go before familiarizing with codebase. Thanks in advance.

[zakwatts]

Hi @PredictiveManish, thanks for showing your interest in this project. Before looking at the codebase, you might find it useful to research a bit about solar forecasts and why they are important for countries to manage to their grids. It also might be useful, depending on your experience in the energy sector, to look into how solar energy is generated and the data factors involved that influence its production (i.e: Amount of sunshine, cloud cover, weather forecasts, solar panel tilt and orientation set ups, dust/pollution factors, Solar inverter types etc). This should help you to understand why we include certain variables and inputs into the solar forecasting model codebase.

[itsbharatj]

Hey @PredictiveManish, welcome to the OCF Community. As a fellow contributor, I can definitely ensure that it is one of the most fulfilling and purposeful organizations.

As @zakwatts recommended, I would suggest you going through the 6-minute talk here: https://www.climatechange.ai/papers/iclr2024/46 to get a brief idea of the need and specifics of the project.

Let me know if you have any questions about the talk.
Looking forward to meaningful contributions, you!

[celestinelindarto]

Hi @zakwatts,

Thank you for organising this project and offering us the opportunity to contribute to TZ-SAM. I am interested in solar forecasting and would like to know further details regarding your project. Please refer to the questions listed below.

• By production phase, do you mean having the forecast integrated within TZ Solar Asset Mapper's web UI?
• Is the coding timeline for this project flexible, i.e. can the timeline be extended to 22 weeks to accommodate other commitments such as work?
• Are potential contributors able to send draft proposals to mentors before the application deadline? This allows us to refine our expectations and obtain feedback before sending in the final proposal.

I look forward to hearing from you.

[zakwatts]

Hey @celestinelindarto, thanks for your interest in this project.

1. Production phase here refers to multiple countries being forecasted for in real time and perhaps served up through an API/UI. This would not be integrated into the TZ Solar Asset Mappers web UI. This would involve having a trained model (or using one of our existing ones), and using live weather forecasts to forecast the PV generation for each country. We have a fair amount of code in place already to help with this (e.g: NWP consumers, APIs ([UK National](https://github.com/openclimatefix/uk-pv-national-gsp-api), sites API example).
2. There is some flexibility around extensions, and this would be dealt with on case by case basis. We understand that Mentees have other work commitments and so try to have some flexibility around this. I think extending a 12 week project to 22 weeks could present some challenges but as mentioned for the right candidate I think this is something that could be discussed.
3. Feel free to ask me questions on this Github discussions thread. I'd be happy to give feedback on here.

[ShivamVashisth28]

Hi @zakwatts and everyone in the community, I'm interested in renewable energy solutions. I’ve been exploring the Open Climate Fix Quartz Solar Forecast repo and learning about the key factors affecting solar energy prediction. This project looks exciting, and I’d love to contribute!

[zakwatts]

Hey @ShivamVashisth28, Thanks for your interest.

[zakwatts]

Hi all @PredictiveManish @celestinelindarto @ShivamVashisth28 @itsbharatj.

I have made some updates to the original brief of the project so it could be worth a re-read when you have a chance. I have suggested that a possible outcome is to forecast all large solar sites globally, using an existing model in the Open Quartz Solar repository. Let me know if you have any questions or other ways you would like to use the dataset! :)

[hrsvrn]

Hi @zakwatts ! I'm interested in contributing to this global solar sites forecasting project. The integration of TZ-SAM's comprehensive solar asset database with the Open Source Quartz Solar Forecast framework presents an excellent opportunity for meaningful contribution. I am particularly interested in the Gradient Boosting Model approach mentioned in the repository and would welcome the opportunity to develop and optimize these forecasting capabilities.

Let me know where to start..

[hrsvrn]

Alright sure!

[hrsvrn]

@zakwatts I have been learning more about how solar forecasting works in the quartz solar project. I have questions about the codebase and some understandings of the project which you may correct if I'm wrong:

• For solar forecasting prediction, you're using PV solar dataset and NWP data to train the model. The readme mentions a YouTube video that also references satellite data. I didn't fully understand the satellite data aspect.

• For solar forecasting predictions, you're currently using a dataset on Hugging Face containing the UK PV solar dataset and using Open Meteo for the NWP data, correct?

• Two models are currently being used for prediction: Gradient Boosting Model and XGBoost. Why not use a Time Series Transformer for predicting? I know there might be compute constraints, but Transformers are probably state-of-the-art models for time series as well. Please correct me if I'm wrong here.

• After my analysis of the TZ-SAM dataset... Is it a PV solar dataset that Open Climate Fix intends to integrate into this project?

[zakwatts]

1. Satellite data is not currently used in the Open Quartz model. We do have a version of this model that does use it, but as live satellite data is not free and publicly available it is not currently used in Open Quartz.

2. The predictions the model is trained off the UK PV solar dataset, to train we used historical forecasts of the NWP which we did not get from open meteo, but in live, we use open meteo to use the latest NWP forecast.

3. We are very open to use TFTs/transformer based approaches, I agree I think it might lead to some accuracy improvements over gradient boosted trees.

4. Correct, we would look to use the information about the solar sites in the TZ-SAM dataset in this project.

[hrsvrn]

@zakwatts thanks for the response once again

1. Ok I see the reason why I didnt see this in the open source repositry
2. So for the TZ-SAM dataset we have to use the global NWP provider right?
3. Should I include that as well in my gsoc proposal? Since I am very eager to work on those as well but since we have to mainly focus on integrating the tz-sam dataset with the nwp global data i will be keeping the transformer implementation a second priorty
4. Alrighttt. So I think I am procedding the correct direction it seems.

[hrsvrn]

@zakwatts any updates on the gsoc and the project in general? Its been two weeks and no communication whatsoever. I am still keen to implement this

[Prxxthxm]

Hi @zakwatts!
I'm intrigued by the project idea and would love to contribute to the same!
Just to check if I've grasped the project correctly, over the course of the project, we are expected to:

1. Run Open Quartz on each of the identified solar farms in the TZ-SAM dataset
2. Aggregate these forecasts and identify trends
3. Utilize this information to generate solar energy production forecasts using predictive weather data, either using Open Quartz or developing an ML model if needed

Please do correct me if I've misunderstood anything, looking forward to working on this!

[zakwatts]

Hey @Prxxthxm, thanks for you questions! Yes, I think you have gauged how this project could develop well! In extension to what you have mentioned some additional tasks could be to:

1. Create a docker container to run the sites so that we could run this live.
2. Create a simple UI, perhaps in Streamlit or Plotly dash to show all of the sites on map?
3. Research other sources of capacities for countries and see if we can apply our learnings from this project.

Do let me know if you have any other questions!

[Prxxthxm]

Thanks a lot for the clarification!
I'll get started on the mentioned steps!

[yuvraajnarula]

Hey @zakwatts zakwatts,

I'm fascinated by your GSOC project using TZ-SAM for global solar forecasting! A few pointed questions:

1. How accurately can satellite imagery predict panel degradation over time, affecting forecast reliability?

2. Are there significant performance differences in forecasting between utility-scale vs distributed solar sites?

3. What's your approach to handling data gaps in regions with less satellite coverage?

After reviewing the project scope, I'd like to suggest three potential ideas, let me know if these align with the project:

1. Develop a comparative benchmark tool that evaluates forecasting accuracy across different climate regions, highlighting where additional data inputs would most improve predictions.

2. Create an uncertainty visualization layer that shows forecast confidence levels by region, helping grid operators prioritize backup resources.

3. Build a simplified forecasting API that allows regional grid planners to query expected output ranges for specific sites during extreme weather events.

I have been actively working with https://github.com/openclimatefix/graph_weather for some time now and have gained a strong understanding of the code. If you have any advice to share, I would appreciate hearing your valuable insights.

The intersection of ML with this comprehensive solar dataset seems incredibly promising for improving renewable energy planning. Looking forward to your thoughts!

[yuvraajnarula]

Hi @zakwatts, I hope you’re doing well! When you have a moment, could you please share your feedback on the approaches. Your insights would be invaluable in helping me create a stronger proposal. Thank you!

[zakwatts]

Hey @yuvraajnarula, Thanks for your interest in this project. To answer your questions:

1. There might be some information in using sat imagery to predict panel degradation. Firstly, it might be tricky, but possible to work out how dirty panels are, which affects their performance. Secondly you can use sat imagery to work out the expected PV generation. Comparing this then to the actual generation can give some insights as to how the panel is degrading over time.

2. Theres some difference between utility scale and household. Often households can be more restricted due to their roofs specific tilt and orientation. This combined, with often poor reporting of specific solar set ups on households such as the tilt and orientation of the panels leads to worsened forecasts. This is why using live data as a feature in the PV helps a lot, to calibrate the forecast to adjust for any inaccuracies in the metadata. Generally, when using the same model, I wouldn't say there are significant performance difference given even/fair set ups for comparisons.

3. Great questions and I don't think its something we have had to deal with yet. We use the Metosat-10 for Europe and Africa and Metosat-9 (IODC) for the work we are doing in India. As we often have quite large archives of the data we can get away with small gaps. If satellites are new and lack training data, I think this is where I could see fine tuning of the existing models particularly useful. Possibly something worth exploring with the existing Open Quartz Model.

[zakwatts]

In response to if your potential ideas. I think they sound interesting. Some thoughts:

1. I think the tricky thing here would be getting access to PV data. The TZ-SAM dataset only records solar sites above a certain size. There are websites that share PV data such as pvoutput.org but most of the sites are households which wouldn't be picked up by TZ-SAM (at least for now, TZ-SAM is looking to expand to include households soon).

2. I think this would be cool to do. Currently Open Quartz does not provide p10/ p90 confidence quantiles, although this would be a nice addition to the model. One way around this could be to make multiple predictions and look at their uncertainty. I've just noticed Open-Meteo publish ensemble NWP data which could be used to work to interpret confidence.

3. This sounds like a good idea.

[yuvraajnarula]

Thank you for clarifying and providing your feedback! I'll begin working on the steps you mentioned.

[codderrrrr]

Hi, @peterdudfield
My name is Saad Waqar and I recently contributed to PVNet repo which got me interested in solar forecasting. I am more than excited to contribute to Open Climate Fix for this year gsoc. I would like to ask:
1- Is there any recommended resources or past projects that could help me get started?
2- Is there any tech stack you want that I should know before project starting?
Looking forward to contributing on this exciting project.

[zakwatts]

Hi @codderrrrr. I think it slightly depends on what project specifically you are most interested in as we have few on!

1. If you interested in this project, I think getting familiar to the Open Quartz repo is a good place to start. Theres a link in the original post on this thread.
2. We're pretty python heavy at OCF. All data prep and ML training is done in python.

[shubhambhar007]

HI @zakwatts ,I am very interested in collaborating with you.
I am currently working at Microsoft as a SDE2.

PFA my proposal

Project Title: Solar Forecast with TZ-SAM: Create Forecasts for Sites or a Grid (Country) Based on Transition Zero’s Solar Asset Mapper

Project Description: The Transition Zero Solar Asset Mapper (TZ-SAM) is a new open-source tool launched by the not-for-profit organization Transition Zero (TZ). It provides a database of approximately 60,000 large solar farms worldwide, identified using advanced AI satellite imagery. This project aims to use the TZ Solar asset mapper dataset to forecast each of the sites it identifies and/or map out total capacities for large sites in each country. These capacities will then be used to generate solar forecasts for different grids around the world.

Motivation: The primary goal of this project is to leverage the TZ-SAM dataset to create accurate solar forecasts, thereby enhancing the ability to predict solar energy production. This will help in better planning and management of solar energy resources, contributing to a more sustainable and efficient energy grid.

Expected Outcomes:

• Basic Solar Forecasts: Generate basic solar forecasts using Open Quartz for every site, grid, or country in the TZ-SAM dataset using a global Numerical Weather Prediction (NWP) provider.
• Production-Ready Implementation: Aim to have the forecasting system in production by the end of the mentoring period.
• Technical Approach:

Data Integration:

Integrate the TZ-SAM dataset with the forecasting system.
Ensure the dataset is pre-processed and cleaned for accurate forecasting.
Solar Forecasting:

Utilize Open Quartz to develop solar forecasting models.
Implement these models to generate forecasts for each site, grid, or country in the TZ-SAM dataset.
Use a global NWP provider to enhance the accuracy of the forecasts.
Dockerization:

Create Docker images for the forecasting system to ensure consistency and ease of deployment.
Use Docker Compose to manage multi-container applications.
CI/CD Integration:

Implement Continuous Integration and Continuous Deployment (CI/CD) pipelines using GitHub Actions.
Automate the testing and deployment process to ensure the system is production-ready.

Timeline:

Community Bonding Period:

Familiarize with the TZ-SAM dataset and existing tools.
Engage with mentors and community members to gather insights and feedback.
Phase 1:

Integrate the TZ-SAM dataset with the forecasting system.
Develop and test the initial solar forecasting models using Open Quartz.
Phase 2:

Dockerize the forecasting system and set up CI/CD pipelines.
Conduct extensive testing and validation of the forecasts.
Phase 3:

Finalize the integration and deployment of the forecasting system.
Perform extensive testing and debugging.
Prepare documentation and tutorials for users.
Deliverables:

Integrated TZ-SAM dataset with the forecasting system.
Solar forecasting models developed using Open Quartz.
Docker images and CI/CD pipelines for the forecasting system.
Comprehensive documentation and user guides.
Regular progress reports and final project report.
Skills Required/Preferred:

Running ML models
Docker, Docker Compose
GitHub Actions
Benefits to the Community:

Enhanced ability to predict solar energy production.
Improved planning and management of solar energy resources.
Contribution to a more sustainable and efficient energy grid.
Future Work:

Explore additional forecasting models and techniques to further enhance accuracy.
Continuously gather user feedback and iterate on the forecasting system to improve its effectiveness.

About Me: I am Shubham Bhardwaj, a Software Engineer with experience in running ML models, Docker, and CI/CD. I am passionate about creating innovative solutions that contribute to sustainability and am excited about the opportunity to contribute to this project through GSoc.
My Linkedin Profile - www.linkedin.com/in/shubham-bhardwaj-ba02b616b

[zakwatts]

Hi @shubhambhar007, thanks for interest in this project. I'm not able to give feedback on individual proposals but if you have any questions about the project or want to discuss ideas around your proposal then feel free to send over any questions.

[yobazy]

Hi Zak,

My name is Bazil, and I’m reaching out to express my interest in the "Solar Forecast with TZ-SAM" project for Google Summer of Code 2025. I was excited to come across this project — it aligns well with both my technical skills and my academic background.

I hold a Master’s of Engineering in Water Resources Management, where I focused on data-driven modeling and machine learning applications. In my current role as a backend and cloud developer, I work extensively with Python and Docker, and I’m excited by the idea of applying those skills to a globally impactful project like solar forecasting at grid scale.

I’ve begun reviewing the linked repositories and reading through the related discussions.

I understand that mentors aren’t able to comment on proposals directly, but if there are any additional resources or project nuances you’d recommend I explore further, I’d greatly appreciate any suggestions.

Thanks again, and I’m looking forward to contributing to Open Climate Fix.

Best regards,
Baz

[zakwatts]

Hi @yobazy, thanks for your interest in this project. I think theres already been some great discussions in this thread around different ideas and resources for the project. Please let me know if you have any further questions or if anything is unclear!

[ghost]

Hi @zakwatts and OCF Team,

I’m Dakshbir Singh, an enthusiastic contributor to Open Climate Fix, and I’m incredibly excited about the "Forecasting Global Solar Sites Using TZ-SAM Data" project for GSoC 2025.I’ve been actively contributing to OCF over the past few weeks, and I’m eager to make more meaningful contributions as I prepare my GSoC 2025 proposal.

I find this project particularly fascinating because using TZ-SAM’s global solar asset data combined with NWP-based forecasts offers immense potential for improving solar energy predictions at scale. The ability to forecast large solar farms across different regions can significantly enhance grid reliability, energy dispatch planning, and sustainability efforts.

After reviewing the project description and previous discussions, I had a few technical questions and would love to get your insights:

1. Ensemble Predictions for Forecast Uncertainty: I noticed that Open Quartz currently generates point forecasts but doesn’t include p10/p90 confidence intervals. Would it be valuable to implement ensemble predictions or use probabilistic models (e.g., GMMs or Bayesian methods) to capture uncertainty? This could enhance the reliability of forecasts, especially in regions with unpredictable cloud cover.
2. TZ-SAM Data Expansion: Since TZ-SAM currently focuses on large-scale solar farms, have you considered integrating smaller solar sites using public PV data platforms like PVOutput.org? Expanding the dataset could improve coverage, particularly in regions with more distributed solar setups.

[zakwatts]

Hi! Thanks for your interest in this project.

1. I like this idea! I think using the ensembles in a clever could allow us to generate some quantiles for the forecast. Alternatively some modifications could be made to the pv-site-prediction repo, which was used to train to the model, to add in the option to predict quantiles as well. https://github.com/openclimatefix/pv-site-prediction

2. We have, there is a lot of useful data on pvoutput.org. However I think for this specific project, the focus should be on using TZ-SAM sites only as PV output only has a very small fraction of actual households. What could be interesting though is to find estimates of the amount of household capacity for every country, or sites below the minimum tz-sam threshold. Perhaps an organisation like https://ember-energy.org/ might have this kind of data?

• On second thought, as we don't have data to validate the model across the world, some data from pvoutput could be quite useful in try to measure forecasting accuracy and how this might vary across different regions.

[Edwardcg17]

Hello @zakwatts and the Open Climate Fix team,

I'm Edward, an electrical engineering computer science student with experience in Python, machine learning, and data processing. I'm very interested in this project and have just submitted my formal application.

I'm particularly intrigued by the potential to use the TZ-SAM dataset for solar forecasting across different regions. A few questions I'm curious about:

1. Is there a specific region or subset of the 60,000 solar farms that would be most valuable to focus on initially for the forecasting model?

2. Are there any existing evaluations of how forecast accuracy varies by geographic region, and would identifying these patterns be valuable to the project?

3. Beyond the Open Source Quartz Solar Forecast, are there other forecasting methodologies you'd be interested in exploring or comparing?

I'm excited about the intersection of ML, climate tech, and energy forecasting in this project. Looking forward to contributing to this important work!

[zakwatts]

Hey @Edwardcg17, thanks for the questions!

1. Good question, I think initially UK assets could be interesting to look at as we know a lot about this region already and might be able to validate some of the results with the data we have. India could also be a great start as we have existing work going on there, and possibly some data to validate results. I think Africa could be a region where this work could have the most impact as well so maybe then Africa or Europe? Then the world.

2. There are some datasets that do exists I have come across:

• https://globalsolaratlas.info/solar-measurement
• https://pvoutput.org/
• https://github.com/Charlie5DH/Solar-Power-Datasets-and-Resources
  This might provide some of the data needed to validate sites across the world. Alternatively, there could be a collaborative part of this by reaching out to specific assets and requesting their live data to validate our model in return for forecasts? I think trying to measure forecasting accuracy across different regions could be tricky, but it would be useful.

3. It would be interesting to compare the two models in Open Quartz and perhaps others out there, the tricky thing again would getting the data to validate the accuracy. We could do this okay in India and the UK but it might be harder for other regions.

[Dakshbir]

Hi @zakwatts and OCF Team,

I’ve been exploring the Open Quartz repo and the TZ-SAM dataset, and I’m genuinely intrigued by the potential of this project to enhance large-scale solar forecasting. A few observations and questions I had:

Model Generalization Across Regions:

Given the variability in meteorological patterns and PV characteristics across regions, are there plans to fine-tune models per region? This could significantly improve forecast accuracy.

Incorporating Cloud Motion Vectors:

Have you considered integrating satellite-based cloud motion vectors into the prediction pipeline?

This could enhance the model’s short-term responsiveness to rapid weather changes, particularly in regions with unpredictable cloud dynamics.

Probabilistic Forecasting:

Would it be beneficial to explore probabilistic forecasts (e.g., GMMs or Bayesian methods) to capture uncertainty?

This could add confidence intervals to the predictions, making them more reliable for grid balancing and dispatch planning.

Looking forward to following this project’s progress and contributing further.

Best,
Dakshbir Singh

[zakwatts]

Hey @Dakshbir, thanks for your questions and interest in this project.

1. Currently out site model is just trained off UK data, there could be some interesting work to fine tune the model to different regions but this would rely on obtaining generation data from those regions, which can be tricky. I'd say if anything this should be considered and extension to this project and not be part of the core goals and outcomes. Although gathering data to quantify the accuracy difference across regions could be useful part of this project once forecasts are set up and running.

2. Lots of our models at Open Climate Fix use satellite imagery to forecast future generation. The Open Quartz model does not currently use this data due to restrictions over using live satellite imagery data. This is not something we are looking to integrate into the model as part of this project. We've recently been creating a model to forecast satellite imagery though, If you interested in learning more about this you can check it out here: https://openclimatefix.org/projects/cloud-forecasting

3. I think it would, some modifications might need to be made to the core model or alternatively NWP ensembles could be used to try to predict quantiles for generation.

[khushi999]

Hello @zakwatts and the Open Climate Fix team,

I'm Khushi Jain, deeply interested in the "Forecasting Global Solar Sites Using TZ-SAM Data" project. As I prepare my proposal for Google Summer of Code 2025, I am particularly curious about the integration of machine learning models with the TZ-SAM dataset.

1. what are specific challenges anticipated in modeling and forecasting across different geographic regions?

2. how do we plan to handle the variability in data quality and completeness from different global solar sites?

3. are there specific model performance metrics that we aim to achieve or improve upon during this project?

Thank you for this exciting opportunity, and I look forward to potentially contributing to this meaningful work.

Best regards,
Khushi

[hughharford]

@zakwatts ha MegaWatts is a great moniker. I'm new to OpenSource, so just saying hi. I read through the many other comments and responses and look to get in a proposal if I can. Thanks for all your answer so far. Cheers for now, Hugh

[harsht24]

Hi @zakwatts, I’m interested in the "Solar Forecast with TZ-SAM" project. I have relevant experience in machine learning and working with satellite data. I know I’m reaching out a bit late, but I’m passionate about the project and will start contributing today.
Looking forward to being involved!

[emlweb]

Google Summer of Code 2025 applications are now closed.

We are currently reviewing all applications. Contributors will be announced 8 May 2025. Thank you!

[peterdudfield]

I'm closing this discussing now as GSOC 2025 is nearly over. Thank you for everyones input and help.

We hope to take part next year and we'll be posting info here