Hi AsterixDB Team and Suryaa, I am writing to express my strong interest in the LLM Agent Protocols/Memory project (Jira: gsoc2026 label) for GSoC 2026. I am also an active open-source contributor in the AI space, with recent merged contributions to the LangChain ecosystem, specifically langchain-google <https://github.com/langchain-ai/langchain-google/pulls?q=is%3Apr+author%3ASYED-M-HUSSAIN+is%3Aclosed> and deepagents <https://github.com/langchain-ai/deepagents/pull/1083>.
I have been following the emergence of the Model Context Protocol (MCP) and find the idea of integrating agentic memory into AsterixDB's persistent storage incredibly compelling. Based on the project requirements, I have started reviewing the AsterixDB architecture and the MCP specification. I am currently: 1. Setting up my local environment to build AsterixDB from source. 2. Drafting a design document following the ASF application template, specifically focusing on how to track query sessions for persistent memory. Since this is a "Major" difficulty project, I want to ensure my initial approach is on the right track. Are there any specific Jira issues or low-hanging fruit bugs I should look at to better familiarize myself with the codebase relevant to this feature? I’m excited to contribute and look forward to your guidance! Best regards, Syed Muhammad Hussain Machine Learning Engineer, Beam AI Research Intern, Empathic Computing Lab – University of Auckland B.Sc. Computer Scientist, Habib University *LinkedIn <https://www.linkedin.com/in/syed-muhammad-hussain-00b2a7214/>* |* <https://github.com/SYED-M-HUSSAIN>**GitHub <https://github.com/SYED-M-HUSSAIN>* |* <https://scholar.google.com/citations?user=TqMFlMYAAAAJ&hl=en>**Google Scholar <https://scholar.google.com/citations?user=TqMFlMYAAAAJ&hl=en>*
