Find open source projects actively accepting contributors. Search repositories, filter by program milestones, difficulty tags, or tech stack.
Use our Orbit AI Matcher to find out! Get instant matching scores based on your developer skills, preferred frameworks, and contribution experience.
Convert your selected open-source project into a winning GSoC, LFX, or Outreachy application using Proposal Studio.
<p>With a surge of machine learning, especially the neural network model which needs to compute a huge amount of gradients, more and more languages' users begin to call for the ability to do high-performance and convenient <em>automatic differentiation</em> (hereafter AD) on functions, whereupon many AD tools are developed, among which is Enzyme. Enzyme is made as a plugin of LLVM and synthesizes differentiated functions directly on optimized LLVM IR code, which provides it with advantages in efficiency and universality. From another aspect, Rust is an emerging high-performance and safe language. Many of its users are trying to bring machine learning to it, thus an high-performance and user-friendly AD tool needed. Enzyme is a good option for this demand. However, without a deep integration into Rust, Enzyme's user experience with Rust is not so good, and more importantly, it may suffer from performance reduction due to wasted Rust meta-information. These problems hamper Enzyme from being widely used in Rust. This project will tackle the aforementioned problems through integrating Enzyme into Rust and then provide high-performance differentiation in Rust.</p>
A multimodal corpus is a treasure trove of language data. In this project, we’re diving into Farsi-language content from public broadcasts. Through the integration of many media formats, a multimodal corpus offers an abundant resource for linguistic analysis, allowing scholars to investigate patterns and behaviors of language across diverse modalities. Using Red Hen’s YouTube pipeline, the initiative aims to build a comprehensive and varied collection of Farsi-language content from public broadcast media. In order to do this, we will use yt-dlp to seamlessly acquire data from YouTube and other media sources, and OpenAI’s Whisper, which is an automatic speech recognition (ASR) system that can generate highquality subtitles for videos, will intervene in the case of videos without subtitles to guarantee that no important linguistic information is lost. One of our main goals is to pinpoint the most effective natural language processing (NLP) pipeline tailored specifically for spoken Persian. Ultimately, the envisioned pipeline will streamline the process, accepting a list of video URLs and seamlessly transforming them into a functional multimodal corpus within CQPweb.
<p>Currently, Atarashi has 4 active agents types that use 9 different types of algorithms to find the similarity between license statements. According to the evaluation of all the agents and their types, it can be seen that there is a need for improvement in the speed as well as the accuracy of the agents. I’ll be implementing the basics of Lucene and percolate search on the top of different agents to increase the speed of the agent without much tweaking with their original algorithm.</p> <p>A python library for code comment extraction is needed to be created with various additional features for providing support to Atarashi as well as FOSSology. This will be packaged & published to PyPI for easy installation and usage.</p> <p>For any Machine Learning/Deep Learning model to be implemented we need a dataset of SPDX open-source licenses. There isn’t any dataset available online so we need to create our own dataset as per the requirements.</p> <p>I’ll be working on the improvement of the semanticTextSim agent and will be introducing a new part of it. Currently, we have doc2vec implementation and I’m planning to introduce the BERT embeddings for finding semantic text similarity.</p>
The Environmental NeTworked Sensor (ENTS) platform is an open-source web portal designed for collecting, visualizing, and analyzing data from large-scale environmental sensor networks. It enables researchers and citizen scientists to monitor sensors measuring soil moisture, temperature, current, and voltage in outdoor settings, supporting critical environmental research. This project aims to enhance the ENTS platform through two key initiatives. The first focuses on system stabilization to address existing visualization and reliability issues, ensuring that data presentation is accurate and consistent. The second initiative will implement a comprehensive Logger Registration System, allowing users to easily add, configure, and monitor their logging devices, with potential integration to The Things Network for LoRaWAN connectivity. By the end of this project, ENTS will feature improved chart synchronization, reliable data point selection, robust error handling, and an intuitive interface for registering and managing logging devices. These enhancements will transform ENTS into a more stable, user-friendly platform that better supports researchers in collecting and analyzing environmental data at scale.
<p>The Heavy Flavour AVeraging (HFLAV) group is responsible for collecting and combining measurements made at different High Energy Physics (HEP) experiments, at CERN and other particle physics laboratories, and combining them using robust statistical procedures.</p> <p>The HFLAV website provides a live snapshot of the latest data obtained from these latest publications. This project aims to improve the experience of users using the HFLAV website. These averages could be of greater value to the particle physics community if the accessibility, interactivity and visibility of the current website can be improved. Thus this project seeks to transform user interaction with HFLAV’s ‘Beauty to Charm’ averages.</p> <p>The plan to increase the accessibility is by redesigning the whole user interface, making it easier to find a relevant data point ( branching fraction) by the use of filters. As the old website is static, thus in order to make the site responsive to the user's input, the website will be refactored and the framework will be changed to Django. The application would be hosted using CERN's OpenShift service instead of the older EOS filesystem it was using.</p>
<p>The Rivet toolkit is a software package for performing data analysis on simulated particle collision events like those in the Large Hadron Collider (LHC). Rivet covers all aspects of collider physics and is the LHC's principle tool for model testing using events simulated with Monte Carlo methods. The current framework for visualizing experimental datasets can be slow and error-prone, and our proposed project will focus on designing and implementing a replacement for this visualization system based on modern rendering technologies while preserving high quality outputs. The goal of this project is to create an intuitive Python plotting package that complements and incorporates a single coherent treatment of the existing classes within the Rivet toolkit. The new visualization architecture will focus on finding a balance between power and simplicity in the user interface, enabling researchers to quickly share their results in beautiful and easily accessible formats. The ability to design interactive web-based visualizations will allow researchers to create engaging and high-quality educational resources for students and individuals interested in physics around the world.</p>
<h3>Improving Discussion Threads and UI/UX of Apache Allura</h3> <h4>Goals</h4> <ul> <li>Improve Discussion Threads (ForgeDiscussion) </li> <li>Improve landing page on installation and general UI improvements and UX bug fixes (Allura)</li> <li>Implementing a Mobile Web View similar to GitHub</li> </ul> <h4>Brief Points</h4> <ul> <li>Implement async loading of comments and allow them to be paginated. </li> <li>Improve emails sent out by the platform, look into how to integrate them with email clients like done by GitHub.</li> <li>Improve viewing attachments, allow embedding some types of attachments. Preview feature.</li> <li>Extend the models in the threads to allow voting(express +1, LGTM etc.) on individual comments. Add support for emoji.</li> <li>Improve the landing page of Forge.</li> <li>Implement a separate Mobile Web View, this involves extending the API and creating new views specifically for mobile web browsers.</li> </ul>
<p>Monumento application at a production ready state for the Android platform. Initially, I had envisioned the project with Flutter completely for cross-platform support as it suited the project idea. But along my way, working out the details and developing the sample prototype for the project idea, I got to know that ‘Landmark Detection’ from the Firebase Vision API in particular wasn’t yet supported with Flutter. To make it work, I needed to write quite good amount of native code for both Android and iOS platforms. If we were use to use on-device processing, that would make the app itself heavy since we would need to train our own custom ML model with a lot of data as an user can try to detect a monument from any angle possible. All these points, got me to reject the idea of Flutter for the complete app development. But at the same time, Flutter really has some great benefits such as extremely good UX/UI development, smooth performance, faster productive development, multiplatform support for future using same codebase. So I thought why not develop a native Android application and integrate Flutter with it for the frontend work and some backend work binding as well if needed.</p>
<p>MIPv6 is the base standard from the IETF for IP based mobility management in wireless networks. The next generation mobility management schemes are mainly following the protocol stack of MIPv6. Until today ns-3 – which aims to replace the aging ns2 and its associated old-featured simulators – does not have even in its latest version a MIPv6 module that is fully compliant to the IETF specification. To fill this gap, I am going to implement a novel MIPv6 simulation model which can be integrated into ns-3. The project idea aims at testing the code in different network scenarios, containing different link layer technologies such as Wi-Fi, WiMAX and LTE. The current implementation of LTE does not have support of IPv6 in ns-3. So, the idea of testing MIPv6 code into LTE would follow IPv6 support implementation in LTE first and then the MIPv6 support. So, ns-3 researchers can benefit themselves with the addition of this feature. Thus, implementation of MIPv6 in ns-3 as the base mobility management solution and providing LTE support within it could help the current network researchers working on ns-3.</p>
<p>In mass spectrometry the isotope pattern of an unknown molecule consists a powerful analytical tool to identify its molecular formula. Modern technological advancements led to a resolution increase in mass spectrometers. This enables mass spectrometers to distinguish parts of a molecule's isotopic fine structure.</p> <p>To determine whether a presumed molecular formula is observed within a mass spectrum the theoretical isotope pattern of the formula needs to be calculated.</p> <h3>OpenMS implements only one theoretical isotope pattern generator that calculates a low resolution version of the isotopic structure. The additional information that a mass spectrometer provides may be absent in a low resolution theoretical isotope pattern and potentially restricts scientists of excluding possible identities of a certain molecule.</h3> <h3>This proposal aims to enrich the OpenMS isotope distribution class with the implementation of three new algorithms for calculating high-resolution isotope patterns.</h3> <p>Ιt proposes an algorithm that belongs in the class of polynomial multiplication methods and two algorithms that consist variations of the Fourier transform methods.</p>
Strong gravitational lenses are rare and scientifically valuable, but finding them in large imaging surveys requires automated methods. Furthermore, simulated data is used due to the lack of available data, but models trained on simulated lensing images suffer a significant performance drop when applied to real observations (the sim-to-real domain gap). This project addresses that gap by combining equivariant neural networks (ENNs), which exploit the rotational and reflectional symmetries of lensing images, with domain adaptation techniques to build a lens-finding pipeline that transfers reliably from simulated training data to real survey images. The project will also develop a hybrid adaptation strategy that chains unsupervised alignment with supervised contaminant rejection to reduce false positives from visually similar objects like spiral galaxies. Through this project, I aim to build a benchmarked ENN+DA pipeline integrated into the DeepLense codebase, a ranked list of lens candidates from an HSC-SSP subregion with catalog cross-matching and property analysis, a cross-survey generalization evaluation, and a final research report with blog post.
<p>The Simple Feature Access Specification (aka. Simple Features) is a common standard for exchanging spatial features that is widely used in geoinformatics in Open Source projects as well as in proprietary programs. It defines a generic data model for storing and accessing geographic data and their geometry types.</p> <p>Protocol Buffers is a format for data serialization which has inbuilt features such as binary serialization, RPC frameworks and IDL. Its unique features such as flexible, efficient, automated mechanism for serializing structured data but with smaller, faster, and simpler manner makes it a very good candidate for data serialization use cases.</p> <p>JTS Topology Suite (JTS), which is a free and open source library containing geometries that correspond to the Simple Feature Access Standard.</p> <p>Main goal of the project is to define and implement of a serialization API based on JTS, which transparently serializes geometries based on the Simple Feature specification using Protocol Buffers. Extending support for other serialization frameworks such as Avro and support for raster data will also be addressed based on time constraints.</p>
Scope: To create a standalone React component from MatchMiner that will use Firebase to provide trial curation across multiple projects uses Requirements: • Familiarize with React, MatchMiner, OncoKB Angular and Firebase Execution • Study previous MatchMiner implementations to leverage that information into the current implementation plan • Refactor Match Miner component • Code formatting • Create standalone component that can be reused across multiple projects • Employ best practice on naming conventions and proper de-structuring Constraints: • Time • Lack of working experience in Angular Risks: • Can only build upon the component, cannot transform or remove features • Unexpected consequences of changes and ensuring failures will be localized to the component Deliverables Task 1: Strip down Match MinerAngular to base component, its most reusable part, via a DRY method. Task 2: Re-factor MatchMiner in React that will be extendable as a stateless component Task 3: Ensure library component in React is developer friendly in order to not burden the End Dev Task 4: Testing in staging environment to ensure proper functionality
<p>Auditory stimuli like music, radio recordings, movie soundtracks or the regular speech are widely used in research. While it is easy for a human to recognize the emotional load of Bach symphonies, how can it be done by a computer? Currently, algorithms are able to analyze low level features like signals energy. Those features are far from capturing how does the stimuli actually sound to us, and the best we can do is to ask a human subject to judge. The following project aims to bridge the gap between human- and machine-like sound understanding by building an audio embedding space. Such embeddings have been proven extremely successful for texts (word2vec) and images (image recognition, convolutional neural networks, vgg). The aim of this project is to offer similar insight into the nature of the audio stimuli. The embedding space will be obtained by a machine learning model trained to perform on various audio task. Deliverables are: the architecture (data loading and preprocessing pipeline, deep neural network for universal feature extraction and classifiers train separately for each task), ready-to-load optimized parameters, and a use case tutorial (github repository).</p>
Today, LLVM's offload compilation keeps host and device code strictly separate; they only reunite late in the pipeline when the device binary is embedded into the host object. By that point, cross-boundary reasoning is largely out of reach. CIR Combine introduces a merge-optimize-split stage into ClangIR's compilation pipeline. After both host and device modules are compiled to CIR, a new driver action merges them into a single heterogeneous translation unit, an offload container holding both modules as nested regions. With both sides visible simultaneously, cross-boundary analysis passes can propagate constants from host launch sites into device kernels, eliminate kernels that are never launched, and infer launch dimensionalities that inform device-side optimization. A subsequent split action extracts the transformed modules and feeds them back into their respective backend pipelines. Deliverables: - Working merge/split skeleton integrated into the new offload driver - Cross-boundary constant propagation pass validated on real CUDA/HIP kernels - Compile-time overhead measurements, and the infrastructure that makes further host-device optimizations possible in the future.
<p>This project aims to take a step towards a solution, which could make LibreOffice able to handle proper table styles. Implementing table styles in Writer is an ongoing effort for a long time now. However, some major steps are already done, like implementing the table style support in the Writer model, or implementing the ODF import and export filters, there is still a lot of work to be done. This project would aim to do the DOCX import part of it.</p> <p>The problem with the current import is that table styles found in DOCX documents are converted into direct formatting at the level of Writer core. This temporarly solution works fine until someone wants to edit the table, which can cause loss of formatting now. Furthermore, doing this conversion upon opening a DOCX document is very slow.</p> <p>The aim of my project would be implementing the DOCX import of table styles and removing the current style-based formatting conversion to direct formatting. This task may cause the need of extending the existing table style features in Writer. For example, an important, not yet supported property is style inheritance.</p>
DeepLenseSim (built on lenstronomy) requires substantial manual intervention for configuring parameters, managing outputs, and iterating on failures across multi-step simulation pipelines, creating bottlenecks in large-scale dataset generation and limiting parameter space exploration. I will build an Agentic AI framework following the HEPTAPOD philosophy to autonomously orchestrate DeepLenseSim workflows across three phases: (1) implementing schema-validated Pydantic tools wrapping the DeepLens class for all four model configurations, an LLM orchestration layer via OpenRouter function calling, human-in-the-loop confirmation flow, and a compatibility layer addressing pyHalo/lenstronomy/numpy version conflicts; (2) building a ParameterScanTool for automated multi-point sweeps (axion mass, halo mass, redshift), a ValidationTool for automated image quality checks, and exploration of RL-based parameter optimization; (3) comprehensive documentation, tutorial notebooks, a test suite with >80% tool function coverage, and integration with downstream DeepLense ML training pipelines via standard export formats (numpy, HDF5, FITS).
The goal of this project is to reintroduce a tree view functionality into Chrome's bookmarks side panel, addressing user feedback and enhancing usability. The current flat list view lacks the ability to view all bookmarks at once, leading to user dissatisfaction. To solve this, we will transform the compact view into a tree view while retaining all new features introduced in the recent UI overhaul. This will involve retaining drag-and-drop functionality for bookmark organization within the tree view, ensuring compatibility with screen readers, and conducting comprehensive testing for usability and accessibility. Key Deliverables: 1. Implement tree view functionality in the bookmarks side panel, allowing users to view all bookmarks at once. 2. Update existing drag and drop functionality for organizing bookmarks within the tree view. 3. Ensure accessibility standards are met, focusing on tab traversal and screen reader support. 4. Conduct thorough testing to ensure the usability, accessibility, and functionality of the tree view. 5. Prepare comprehensive documentation detailing the implementation, usage, and features of the new tree view in Chrome's bookmarks side panel.
This project focuses on systematically expanding regression test coverage for the ESP codebase by targeting high-risk, untested logic paths rather than merely increasing overall coverage percentage. Through detailed codebase exploration, critical gaps have been identified in components such as lottery assignment logic, class registration flows, and mailman integrations. These areas involve complex logic, invariants, and edge cases that currently lack sufficient test coverage and pose a high risk of silent regressions. The project introduces invariant-based testing for the lottery system to ensure correctness properties such as no conflicts, no overcapacity, and valid grade constraints across arbitrary inputs. Additionally, integration tests will be implemented for mailman functionality using mocking to validate subprocess interactions without external dependencies. A reusable testing infrastructure will be developed to support scalable regression testing, along with targeted performance checks to detect potential degradations in critical views. The outcome will be a robust regression testing framework that transforms fragile, untested paths into enforceable guarantees, significantly improving system reliability and maintainability.
<h3>Visualisation of the MOLTO-3BP Incorporating Matlab Python SDK</h3> <p>I am aware there are many applicants, thus I sincerely apologize and regret my delay but hope I can be given an assignment/evaluation test.</p> <p>I'm applying to the track for the visualisation of the MOLTO-3BP network as I am enrolled to start a graduate program in CS using distributed technology and am interested in the applications of orbital equations and computing problems.</p> <p>My motivation lies in my current interest for Backend and Frontend applications as I also work with a company managing their ML algorithms for data prediction in arbitrage contexts as lead analyst and developer for the Full-stack infrastructure to support such volume and structure the visualisation process for qualitative information (as quantitative is given by volume in this case). I am now interested in leveraging such skillset to perhaps a more researching-focused world in lieu of an economic speculative industrial motivator.</p> <p>I believe the 3BP is very interesting from a theoretical point of view and can be greatly improved with visual filter and newer GUI revamp to its already robust computational engine (orbital eq in the halo numerical model)</p>
<p>Deep Learning Networks have proven to perform well for a wide range of problems, especially those requiring large labelled dataset to learn patterns. A new algorithm that has taken the Deep Learning research community by a storm is that of Generative Adversarial Networks (GANs) introduced by Ian Goodfellow in 2015. This algorithm has a lot of potential owing to the fact that it can generate data that is quite similar to the data given for learning in addition to faster generation of samples when compared to fully visible belief nets. It therefore makes GANs immensely useful in simulation of particle physics and astrophysical data.</p> <p>Since ROOT is a data analysis tool extensively used for applications in particle physics and features a dedicated machine learning submodule, Toolkit for Multivariate Analysis (TMVA), it is essential to include a GAN implementation in the toolkit.</p> <p>My project would focus on integrating an optimized GAN implementation in the TMVA DNN library with the help of already existing implementations of Deep Network Models. It would also involve enabling GPU Implementation of GANs using Nvidia’s CUDA library.</p>
<p>Design and implementation of digital filters is essential for electronics engineers. Digital filter design in hardware usually is a two stage process; algorithm development in a language like Python, Matlab or Java and RTL design in a Hardware Description Language like VHDL or Verilog. myHDL is a Python module for developing, synthesizing and testing HDL code. PyFDA is a GUI based tool in written in Python/Qt for analysing and designing discrete time filters. This project aims to leverage and demonstrate the advantages of myHDL, PyFDA and python in general in the field of Digital Filter design. Algorithm exploration, modeling, designing and simulating of digital filters can all be done in Python and PyFDA. The project will focus on development and rigorous testing of digital filter designs and their integration with PyFDA. The algorithms for each type of filter will be chosen such that the hardware utilization is kept to a minimum. Initially Systolic FIR filter and parallel path IIR filter will be implemented. Later more complicated implementations of non-linear estimation filters will be explored. The filters will be rigorously tested and integrated with PyFDA.</p>
This project aims to enhance the existing appointments calendar in OpenMRS by transforming it into a more interactive, flexible, and user-friendly scheduling system. The current implementation provides only a basic monthly view with limited details, which is insufficient for real-world clinical workflows. The proposed solution introduces multi-view support (monthly, weekly, daily), enabling users to visualize schedules at different levels of detail. It will display detailed appointment information such as patient names, time slots, and statuses directly within the calendar. Additionally, navigation-based interactions will be replaced with modal-based views to preserve user context and improve usability. The project also includes support for multiple calendar systems (Gregorian, Islamic, Ethiopic, Persian) to ensure global adaptability. The implementation will follow a modular architecture using React microfrontends, OpenMRS REST APIs, and optimized data-fetching strategies. Deliverables include: Multi-view calendar (monthly, weekly, daily) Detailed appointment rendering Modal-based interaction system Multi-calendar system support Performance optimizations and testing
I propose to enhance the Mifos Initiative's Kotlin Multiplatform (KMP) Multi-Module Project Generator to streamline cross-platform application development and make it more valuable for developers. My plan is to create a customizable Material Design 3 component layer for a consistent UI across platforms, implement a unified configuration system, and enhance GitHub Actions workflows for comprehensive CI/CD. I'll develop pre-built infrastructure components (networking, database, storage), streamline dependency management with Gradle plugins, improve the sync directories system, add a type-safe API client with Ktorfit, and automate library publishing to Maven Central. With my experience in Kotlin (3+ years) and Kotlin Multiplatform (2+ years), plus my extensive contributions to Mifos projects, I have the technical foundation to implement these enhancements successfully. I've created a detailed 15-week implementation plan with clear deliverables for each stage. These improvements will transform the KMP Project Generator into a more powerful and user-friendly tool, benefiting both the Mifos ecosystem and the broader KMP community by reducing development time and ensuring best practices across platforms.