Skip to content
View Alexanderk30's full-sized avatar

Highlights

  • Pro

Block or report Alexanderk30

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Alexanderk30/README.md

Hi, I'm Alex Kimmig 👋

CS @ UGA '26 · Athens, GA · Targeting FDE / Applied AI roles

LinkedIn

What I Build

I design and ship agentic AI systems : multi-agent pipelines, MCP infrastructure, and production grade tooling built on the Anthropic SDK. My focus is on the intersection of real world deployment and frontier model capabilities.

Featured Projects

Project Description Stack
context-override-resistance Eval environment reproducing "acknowledged-but-stale" tool-call failures: model commits to a correction, then acts on stale context anyway. 655 trials across Haiku 4.5, Sonnet 4.6, GPT-5.4, Gemini 3.1 Pro. Python
ClaudeMCP Multi-tenant MCP proxy gateway with per-tenant auth, ACLs, rate limiting, and 80+ tests Python
agent-mesh Multi-agent orchestration runtime with typed message passing and supervisor routing Python
mcp-registry Self-hosted MCP server registry with tool discovery, health checks, and metrics dashboard TypeScript
eval-harness YAML-based pipeline for evaluating multi-step agentic Claude workflows with a judge model Python
research-agent Self-hostable deep research pipeline with parallel search, deduplication, and cited synthesis Python

Tech & Tools

Python TypeScript Anthropic SDK MCP Multi-agent Systems RAG FastAPI ChromaDB Git

Let's Connect

Pinned Loading

  1. ClaudeMCP ClaudeMCP Public

    Multi-tenant MCP proxy gateway — per-tenant auth, glob-pattern ACLs, rate limiting, hot-reload config, and 80+ unit/integration tests

    Python

  2. OpenclawDerivativeTrading OpenclawDerivativeTrading Public

    Multi-agent options trading system — 8 coordinated AI agents executing Iron Condor, Credit Spread, and Wheel strategies via Alpaca with ML signal validation and risk management

    Python 1

  3. quickclaw quickclaw Public

    Multi-agent options trading system — 8 coordinated AI agents executing Iron Condor, Credit Spread, and Wheel strategies via Alpaca with ML signal validation and risk management

    TypeScript

  4. context-override-resistance context-override-resistance Public

    RL-style eval measuring intent/action divergence in frontier agents: model acknowledges a correction, then acts on the stale value anyway. 3 scenarios, 371 trials on claude-haiku-4-5, Sonnet 4.6, G…

    Python