New Contributor Intro: Vedit Agrawal (@BugHunterX2101) | Looking to pair with a MeshMate for Meshery UI & AI Adapter

Community
, , , ,
1

Hi all, I’m Vedit Agrawal (@BugHunterX2101 on GitHub), a final-year CS Undergrad and an Oracle Cloud Certified Data Science & Generative AI Professional. I’m joining Layer5 to contribute to Meshery’s core repositories and looking to pair with a MeshMate.

My engineering background spans production AI infrastructure, distributed systems, and full-stack TypeScript/Python/C++:

  • ScaleDux (Applied AI Intern): Cut production API latency by 40% by refactoring synchronous FastAPI inference into asynchronous Celery and Redis worker queues monitored via Prometheus and Grafana; improved XGBoost/scikit-learn accuracy by 18% through Pandas/NumPy feature pipelines; and automated containerized YOLOv8/TensorFlow deployments on AWS EC2 and Lambda via GitHub Actions.

  • Systems & Full-Stack Projects: Engineered Sentinel Gateway (Next.js, TypeScript, Drizzle ORM, Neon PostgreSQL) with a SENSE-DECIDE-ACT-EXPLAIN loop, 1.5s SSE p99 telemetry, blast-radius estimation, and zero-restart circuit breaking; built a Jenkins Log Intelligence Engine running 3-attempt Celery/Redis TF-IDF retrieval over 10 MB build logs with Llama 3.3 70B for root-cause diagnosis; created a 24-language Enterprise Privacy Suite using RapidOCR bounding-box geometry, Presidio, spaCy NER, and Auth0 RS256 JWKS caching; and built Quantum Sentinel, accelerating Level-2 order-book simulation in C++ via pybind11.

  • Multimodal Edge Research: Designed a 2.5M-parameter network (eGeMAPS, BiLSTM, MobileNetV3, adaptive attention for missing-modality drops) achieving 94.2% attack detection and 89.4% emotion classification accuracy across SWaT, IEMOCAP, and AMI inside 10.2 MB / 2 GB VRAM at 70 ms latency.

I’ve explored the Meshery Playground and local setup, and I want to focus my contributions on:

  1. Meshery UI (React / Next.js / TypeScript): State handling, schema validation forms, and visual components.

  2. AI/LLM Adapter (#20994) & MCP Server (#19446): Provider connection lifecycle, credential isolation, tool definitions, and token-budgeted context assembly.

I’d appreciate being paired with a MeshMate to review my dev setup and help point me toward an open issue in Meshery UI or the AI adapter tracks to submit my first PR. Looking forward to meeting everyone on the Thursday Newcomers call!

2

Hi Vedit, welcome to the Meshery community! :tada: Impressive background — and great that you’ve already explored the Playground and local setup.

For the Meshery UI and AI/LLM adapter tracks, the contributing guides at docs.meshery.io are a good next step for finding open issues to start with, and the Thursday newcomers call is a great place to have your dev setup reviewed and meet MeshMates who work in those areas. See you there!