Fetching the latest programs, projects, and workspace data.
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.
Data Scarcity is a Huge and Evergrowing Problem as new tasks and problems are tackled. Sometimes collecting substantial amount of data is almost impossible/ extremely expensive as well incases of Low Resource Languages etc. hence Deep Learning Practitioners have moved to Data Augmentation Techniques to create synthetic data from real data. In Computer Vision the parallels are very clear for instance zooming in on an image or rotating an image retains labels in most cases however in Language wrong modifications can lead to mislabeling of data pretty quickly and hence it's much more trickier to perform Data Augmentation in these cases. Through my proposal I would like to add 10 Data Augmentation Techniques most of which are Rule Based but have been used extensively in NLP literature to achieve substantial improvements. Having these Techniques as layers would make it convenient for users to add these techniques to their workflow and push models into training. Tweaking augmentations would also be less of a hassle as currently most Data Augmentation libraries don't offer seamless integration with models without writing custom code. \ The 10 Techniques I plan to implement are Synonym Replacement, Random Deletion, Random Swaps, Random Insertion, Stop Word Dropout, Character Level Swap, Keyboard Typo, SentenceShuffle, Remove Duplicate Sentences and Back Translation All of these techniques have been used in Heavily Cited Works making them the perfect candidates to add to the list. By the end of the GSOC tenure my goal is to have these techniques fully implemented as layers with accompanying tutorials to get started with them for new users to quickly add these to their workflows once the library has a formal release.
Deploying AWS Lambda functions using AWS SAM (Serverless Application Model) requires manual creation of a deployment descriptor, posing challenges for Swift developers due to learning a new DSL and adopting new tools like SAM CLI and Docker. To simplify this process, the community proposed a Swift-based DSL and a Swift package plugin. However, these solutions rely heavily on accurately mimicking the evolving SAM AWS deployment descriptor. To address this, this project proposes to automatically generate the AWSLambdaDeploymentDescriptor library based on the SAM template definition. That generated code is used by the Swift-based DSL to create SAM YAML templates required to deploy Lambda functions and their dependencies to AWS.
Symbolic regression (SR) aims to discover human-interpretable mathematical expressions from numerical data. While recent transformer models achieve high token-level accuracy on physics datasets, they struggle to recover exact, functionally correct equations. Conversely, generative techniques like Genetic Programming (GP) excel at numeric fitting but explore expression space blindly. This project solves this structural gap by combining modern transformer priors with generative search. I will build upon previous ML4SCI projects to implement a hybrid pipeline: an xVal-encoded seq2seq transformer generates structurally sound candidate expressions, which seed a GP population. The GP outputs are then ranked and used to iteratively fine-tune the transformer via Direct Preference Optimization (DPO).
This project will create a Kubernetes deployment for the full stack: the OpenTelemetry Demo application, the Jaeger backend components (Collector, Query), and OpenSearch as a storage backend. The environment will be automatically redeployed weekly to ensure it is always fresh and to solve the problem of data retention. The entire stack will be hosted on a managed Kubernetes cluster (Oracle Kubernetes Engine) within an Oracle Cloud environment generously donated to the project. Expected Outcome: 1. A working, publicly accessible Jaeger demo featuring the OpenTelemetry Demo application, deployed on Oracle Cloud. 2. Fully automated, weekly deployments using the existing Helm- based automation and GitHub Actions. 3. The entire environment defined as "Configuration as Code". 4. A secure deployment following best practices for public-facing services. 5. The UIs for Jaeger, the OTel Demo, the load generator, and OpenSearch Dashboards exposed -publicly via the existing demo.jaegertracing.io URL. 6. Public-facing documentation on the Jaeger website explaining the demo architecture and linking to the automation code.
Current methods in the DeepChem library struggle to understand complex polymers, limiting their application in drug delivery and materials science. This project proposes a solution: two new representation methods for these molecules. One utilizes a detailed graph structure, capturing the intricate connections within the polymer, while the other leverages a special text format (BigSMILES) for efficient storage and retrieval. By implementing these methods, DeepChem will be significantly better equipped to handle these repetitive and complex molecules. To achieve this, the project will deliver a comprehensive toolkit: new graph-based and BigSMILES featurizers for polymer data, code for data loaders and integration with DeepChem's MoleculeNet framework, trained and evaluated GCN and MAT models utilizing the new representations, and finally, tutorials, documentation, and a public dataset of polymers represented using the new methods – all to empower the scientific community for further research in this field. The proposer has experience in relevant areas like machine learning and working with polymers.
PR_SET_VMA mechanism is used heavily in Android for naming memory regions. It is passed to the prctl(2) system call along with the range of addresses to be named. The name appears in /proc/pid/maps and can be used to identify memory regions for debugging and accounting purposes. The patch has been carried in the Android tree for several years. The patch needs to be adapted to mainline, improved and finally upstreamed. This project will give the mentee a good understanding of virtual memory area structures in the Linux kernel as the name for the memory region is stored in these structures and there are various cases where a memory map needs to be split or merged depending on its name compared to its immediate neighbors.
<p>Currently the software uses a lot of exteranal libraries like HTML2PDF, MPDF, FPDF and TCPDF. My aim for the project is to make a custom PHP library based on TCPDF as backbone to create PDFs autonomously according to the data present in the database and the choices made by the user.</p>
Parameterized tests allow the test logic and the data passed into the test to be specified separately. Such a feature would have wide applicability in Linux kernel testing. This project would entail designing a new API for defining parameterized tests, integrating parameterized tests into the KUnit reporting mechanism and converting some tests over to use this new API. We envision something similar to https://dzone.com/articles/junit-parameterized-test Adding the feature makes it easier to write Kunit tests that need to iterate over a set of values without duplicated code and with clear reporting of failures.
<p>Security Assertion Markup Language (SAML) is an XML based security token which is used to exchange authentication and authorization data between two entities. It is one of the most widely used open standards to perform brokered authentication and web browser based single-sign-on. "Assertion Query/Request Profile" defines a protocol for requesting dynamic or existing assertions by reference or by querying on the basis of a subject and additional statement-specific criteria. Assertion Query/Request Profile defines five major request/query message types which can be used to request assertions from SAML authority.</p> <ol> <li>AssertionIDRequest</li> <li>SubjectQuery</li> <li>AuthnQuery</li> <li>AttributeQuery</li> <li>AuthzDecisionQuery</li> </ol>
<p>Hyperledger Fabric has been available for several years and there are many production networks using Chaincode. Fabric-X has just been released and it introduces changes to the current way of deploying applications. This brings up the question on how to migrate existing applications and chaincodes to this new environment.</p><p>The goal of this mentorship project is to:</p><ul><li>Explore the feasibility to run existing Fabric Chaincode on Fabric-x;</li><li>Design and develop a proof of concept that showcases Go Chaincode for Fabric-X;</li><li>Implement a prototype of the components identified;</li><li>Identify a demo use case (i.e., Fabric Samples, CC-Tools, Fabric Private Chaincode) that comes with go chaincode;</li><li>Produce documentation that explains the architecture and how it can be used as well as potential avenues of development.</li></ul><h3>Learning Objectives</h3><ul><li>Deep dive into Hyperledger Fabric and Fabric-X.</li><li>Experience an active open-source community.</li><li>Mastering open-source workflow.</li><li>Enhance your designing and implementing skills, testing, documentation, and project management.</li></ul><h3>Expected Outcome and Deliverables</h3><ul><li>Proof of Concept that showcases an alternative on how to migrate classic Fabric go chaincode to Fabric-X.</li><li>Documentation that explains the demo and how to run it.</li><li>Bonus: a technical blog post and meetup talk.</li><li>Having fun.</li></ul><p><br></p><p><br></p><p>Lean more at <a href="https://github.com/LF-Decentralized-Trust-Mentorships/mentorship-program/issues/59" rel="noopener noreferrer" target="_blank">https://github.com/LF-Decentralized-Trust-Mentorships/mentorship-program/issues/59</a></p>
The goal of this project is to enhance the user experience of gccrs by enabling it to emit error codes similar to rustc. This will help bridge the gap between the two test suites and eventually enable the rustc testsuite to run on gccrs. The project involves implementing necessary changes in gccrs to emit rustc error codes throughout its frontend code. Additionally, it entails expanding the code responsible for error emission to incorporate greater functionality. Furthermore, a comprehensive comparison between the rustc and gccrs test suites will be conducted to improve stability. The ultimate outcome will be an improved user experience for gccrs users, with an increased number of error emissions in various contexts and more informative hints on code resolution. By bringing gccrs closer to parity with rustc, this project aims to enhance its accessibility for users and developers alike.
Recently, research has been made into the process in which we could tune an existing generalized LLM (such as ChatGPT 3.5, 4) for the purposes of generating HDL (especially Verilog/Chisel) code from English. However, we still need a method of evaluating these models so that we can assure that they have full coverage over the language. To this effect, we create multiple large benchmarks in both Verilog and Chisel along with a respective set of prompts. In addition, we produce a program to test the syntax, functionality, and quality of the generated code against our benchmarks.
<p>The aim of the project is to enhance the representation of the wire frame views by adding annotations to the sketch.</p>
<p>Image captions are a small piece of text, usually of one line, added to the metadata of images to provide a brief summary of the scenery in the image. It helps text based Information Retrieval(IR) systems to "understand" the scenery in images. It is a very useful feature, yet a challenging and interesting problem in the domain of computer vision.</p> <p>The objective of this project is providing Apache Tika, image captioning capabilities and a scalable architecture to deal with deep learning models in the future.</p>
<p>The Real-Time Executive for Multiprocessor Systems (RTEMS) is an open-source real-time operating system (RTOS). In the current RTEMS version, a lower-priority ready task must wait if all the processors included in its affinity mask are executing higher-priority tasks. Since a lower-priority task can never “dislodge” a higher-priority task that could also execute elsewhere, this may needlessly prevent some tasks from being scheduled, even if some processors idle as a result[1].</p> <p>This project aims to add the Strong Arbitrary Processor Affinity (Strong APA) scheduler to RTEMS. Strong APA scheduler would allow higher-priority tasks to be ”dislodged” or moved among processors in order to make space for lower priority tasks that are limited by affinity constraints [1]. Consequently, this would allow RTEMS to achieve improved schedulability (i.e., lower response-time bounds).</p> <p>References : [1] Cerqueira, Felipe & Gujarati, Arpan & Brandenburg, Bjorn. (2015). Linux's Processor Affinity API, Refined: Shifting Real-Time Tasks Towards Higher Schedulability. Proceedings - Real-Time Systems Symposium. 2015. 249-259. 10.1109/RTSS.2014.29</p>
<p>Minimalist 3D scan (taking multiple pictures, recording camera position, using opendrone map to rebuild the 3D scene) with the house faces (accessible/visible faces) then load it as georeferenced data with the database and being able to export the data back on MapMint for 3D viewing.</p>
When using CoreDNS to serve DNS traffic over TLS/HTTPS certificates have to be managed (created and renewed) manually. The goal of my project is to automate this task, using the ACME protocol. If successful, CoreDNS can be hosted on a domain, such as dns.example.com, and automatically obtain a valid certificate (for example from Let's Encrypt) and it will also automatically be renewed before it expires, reducing toil for CoreDNS operators.
<p>This project aims to improve backend test coverage to 100%, and then migrate the backend codebase to be simultaneously compatible with both Python 2 and Python 3, while putting measures in place (like lint checks) to ensure that the backend code always remains compatible with both python 2 and python 3, regardless of subsequent developer changes. The reason these two projects are linked is because one prerequisite for a safe migration is to have full test coverage, so it's important to make sure that the backend coverage is 100% before migrating.The project would then make sure that all libraries Oppia uses are compatible with python 3. This project would also standardize all scripts in the codebase to be written in Python (currently, there is a mixture of bash and Python being used). The project would end with creating a small list of remaining steps that need to be taken for a final migration to python 3 (once a solution is found for the GAE dependency issues).</p>
In particle physics, a cross section is a measure of the likelihood that particles will interact or scatter with one another when they collide. It is a fundamental quantity and its computation can become time intensive. This project aims to apply state-of-the-art machine learning techniques, particularly transformer-based models, for the generation of symbolic representations of the squared amplitudes in HEP. The proposed sequence-to-sequence system aims to improve upon the existing time-intensive processes of calculations of the squared amplitude . This project builds upon the previous work which includes vanilla models, and extends it further to advanced transformer models.
*OPEN TO US BASED STUDENTS ONLY* - Accepting applications until May 23, 2025 Currently, SONiC community testbed use vendor NOS (cEOS) to create community topology. This puts unnecessary dependency for people to setup community testbed. SONiC has cSONiC image for a long time. The project here is to incorporate cSONIC into SONiC community testbed. This will be much easier for people setup community testbed. Internship Objectives: - Setup cSONiC as neighbor devices in sonic community testbed. - Enable warm reboot LACP lag extension for cSONiC neigbhor. - Enable Macsec for cSONiC neighbor. Expected Outcomes and Deliverables: - Complete the development and setup the community testbed with cSONIC as neighbor. - Submit a pull request and upstream the code changes to the SONiC project. Recommended Qualifications: Education: Current undergraduate student. Skills & Interests: - Eager to contribute to the SONiC open-source community. - Have python/ansible experience. - Academic background in Computer Science, Computer Engineering, or a related field.
RTEMS currently lacks C11 Annex K bounds-checking functions (strcpy_s, memcpy_s, sprintf_s and ~45 others) which provide runtime buffer-size validation critical for safety-critical embedded systems. Without these, RTEMS developers writing software for spacecraft, medical devices, and automotive controllers are forced to use unsafe standard functions or write their own wrappers — neither is acceptable for safety-critical code. This project packages safeclib (rurban/safeclib) through the RTEMS Source Builder (RSB) so any RTEMS BSP can access these functions via -lsafec, directly strengthening memory safety across all RTEMS applications. The approach uses pkg-config for BSP-specific flag detection and librtemsdefaultconfig.a for configure-time link tests. Deliverables include a working RSB recipe, full test suite validation (~126 tests) under QEMU SPARC emulation, all ~45 Annex K functions documented in the RTEMS POSIX Compliance Guide, and a release notes entry.
PyMC uses PyTensor as its computational backend, which allows efficient optimisation, rewrites and evaluation of mathematical operations. Using a recently introduced framework (CoLa- Compositional Linear Algebra), we identify a lot of algebraic simplifications and implement them as graph rewrites. As a result, we would be speeding up multiple model families in PyMC, such as Time Series, State-Space models and Gaussian Processes. The goal is to simplify efforts in implementations while achieving a high degree of computational efficiency
<p>Image Sequencer is a sequential image processing library completely written in javascript which currently supports Node.js and the browser. It has the capability of outputting the processed image at each step as well as a CLI for ease of use right from the terminal. Goal is to enhance Image-Sequencer for both users and contributors by adding more modules, enhancing existing ones, improvisations on demo, an improved CLI and a plugins system to make module contribution easier and independent.</p>
<p>Over the last ten years, Mozilla has been using Buildbot as its primary continuous integration tool in order to schedule builds and tests for every push that is sent to Treeherder for evaluation. As Mozilla is slowly moving towards using TaskCluster as its primary continuous integration system, an important step would be make Treeherder, the primary UI used for handling builds and tests, compatible with TaskCluster. The current Treeherder interface allows us to schedule buildbot jobs. This proposal aims to integrate Treeherder's job scheduling user interface with TaskCluster jobs. While doing this, it hopes to improve the overall integration of Treeherder with TaskCluster.</p>