Skip to content
View cornelcroi's full-sized avatar

Block or report cornelcroi

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
cornelcroi/README.md

⭐ Project Highlights

Project Description
πŸ€– Agent Squad Multi-agent orchestration framework for building complex LLM-powered AI systems (7k⭐)
🧩 Context Lens MCP server enabling AI assistants to search local files and GitHub repositories
πŸ“Š Data Lens Ask questions about spreadsheets using natural language and LLM reasoning
πŸ€– Ask James Tool for querying multiple LLMs for second-opinion reasoning
🌍 Street Lens AI-powered travel platform generating audio guides from LLM pipelines

πŸš€ About Me

I build AI systems and developer tools powered by large language models.

Staff GenAI Solutions Architect at Betclic and former AWS architect working on GenAI prototypes across EMEA.

My work focuses on:

β€’ multi-agent systems β€’ LLM developer tooling β€’ MCP servers for AI assistants β€’ serverless AI architectures

Author or co-author of several open-source projects including Agent Squad (7k⭐).

Previously published on the AWS Blog and community.aws.

Always building tools that make AI systems easier to design, test, and operate.

πŸ€– AI Systems & Tools

πŸ€– Agent Squad (7k⭐)

Multi-agent orchestration framework for building complex LLM-powered AI systems.

β€’ 7k+ stars – #5 most popular AWS Labs project β€’ Supports Python and TypeScript β€’ Used to coordinate specialized AI agents

Role: Co-author

Links: β€’ GitHub β€’ NPM β€’ PyPI


🌍 Street Lens

AI-powered travel discovery platform generating audio guides for places around you.

β€’ GenAI pipeline transforming structured place data into narrative audio guides β€’ Scalable AI-generated content architecture β€’ In production: streetlensapp.com | App Store

Role: Co-Founder


🧩 MCP Servers for AI Assistants

Tools enabling AI assistants to interact with real-world data.

β€’ Context Lens – semantic search over local files and GitHub repositories https://github.com/cornelcroi/context-lens

β€’ Ask James – get a second opinion from another LLM inside your assistant https://github.com/cornelcroi/ask-james

β€’ Data Lens – query spreadsheets using natural language (Excel, CSV, Parquet) powered by DuckDB https://github.com/cornelcroi/data-lens

☁️ Infrastructure & Systems

πŸš€ CloudFront Hosting Toolkit

Open-source CLI to deploy fast and secure frontends using Amazon CloudFront.


🍽️ Food Analyzer App

GenAI-based nutritional web app for analyzing shopping and cooking recipes. Built during an AWS hackathon, then demoed at AWS Summits worldwide as a showcase for GenAI capabilities.

  • Role: Co-Creator
  • Links: GitHub

πŸ”€ A/B Testing at the Edge for E-commerce

A/B testing at the edge using Amazon CloudFront for personalized content at scale.

  • Role: Owner & Maintainer
  • Links: GitHub

πŸ”’ Secure Media Delivery at the Edge

Solution for protecting premium video content delivered through Amazon CloudFront. Now maintained by the AWS Solutions team.

  • Role: Creator (sole coder within a team)
  • Links: GitHub

✍️ Technical Writing

Date Title Platform
12 NOV 2025 Context-Lens: A Serverless, Open-Source MCP Server for AI Document Understanding Blog
28 NOV 2024 Unlock Bedrock InvokeInlineAgent API's Hidden Potential with Multi-Agent Orchestrator Blog
12 SEPT 2024 Beyond Auto-Replies: Building an AI-Powered E-commerce Support System Blog
04 JUN 2024 Introducing CloudFront Hosting Toolkit AWS Blog
10 JAN 2023 How DAZN Uses AWS Step Functions to Orchestrate Event-Based Video Streaming at Scale AWS Blog

πŸ”¬ Current Interests

β€’ multi-agent architectures β€’ LLM evaluation and benchmarking β€’ AI developer tooling β€’ MCP ecosystem β€’ AI system observability

πŸŽ“ Workshops

  1. πŸ”¬ A/B testing at the edge
  2. πŸ’° Scaling cost effective architectures

Always building.
If you're working on interesting AI systems or developer tools, feel free to reach out.

LinkedIn

Pinned Loading

  1. awslabs/agent-squad awslabs/agent-squad Public

    Flexible and powerful framework for managing multiple AI agents and handling complex conversations

    Python 7.5k 709

  2. context-lens context-lens Public

    Semantic search knowledge base for MCP-enabled AI assistants. Index local files or GitHub repos, query with natural language. Built on LanceDB vector storage. Works with Claude Desktop, Cursor, and…

    Python 20 3

  3. data-lens data-lens Public

    Ask questions about spreadsheets in plain English. Excel, CSV, Parquet analysis powered by DuckDB. Natural language to SQL for MCP-enabled AI assistants. Works with Claude Desktop, Cursor, and othe…

    Python 2