Skip to content
View saschabuehrle's full-sized avatar

Highlights

  • Pro

Block or report saschabuehrle

Report abuse

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

Report abuse
saschabuehrle/README.md

Hey, I'm Sascha

AI x Technology x Product leader, 2x founder, and open-source contributor based in Switzerland.

I've spent 10+ years building and scaling high-performing teams at the intersection of AI, product, and technology — from early-stage founding through 30-person orgs across the EU, CH, and US.

Currently focused on AI agent infrastructure and open-source tooling.


What I'm Building

cascadeflow — An open-source AI agent runtime intelligence layer (Python & TypeScript). Optimizes cost, latency, quality, and compliance inside agent execution loops through speculative cascading. 40–85% cost reduction, sub-5ms overhead.

Integrates with: n8n · LangChain · OpenAI Agents SDK · CrewAI · Google ADK · Vercel AI SDK · Ollama · vLLM

@cascadeflow/n8n-nodes-cascadeflow — Official n8n community nodes for cascadeflow, bringing intelligent model cascading to workflow automation.


Contributor

Project Stars
PrefectHQ/fastmcp GitHub stars
aden-hive/hive GitHub stars
raycast/extensions GitHub stars
neomjs/neo GitHub stars
generalaction/emdash GitHub stars
aliasvault/aliasvault GitHub stars
can1357/oh-my-pi GitHub stars

Background

Lemony.ai Co-Founder & CEO — Built 16-person team, enterprise AI platform, 300+ orgs
JetBrains Entrepreneur in Residence — Future of developer tooling, AI prototyping
Piavita AG CPTO — Scaled to 30 engineers, $17M raised, MedTech IoT, Forbes featured
Techniplas Embedded Systems — BMW, Mercedes, Rinspeed concept cars, 100K+ units
Bosch Talent Program — Bluetooth mesh, embedded sensors, IoT protocols

Tech I Work With

Languages TypeScript · Python · Node.js · C/C++ · CUDA · SQL AI/ML LangChain · OpenAI SDK · SageMaker · Ollama · vLLM · CUDA Cloud AWS · Kubernetes · Docker · Terraform · GitHub Actions AI Tools Claude Code · Codex · Cursor 

Homelab

Running NVIDIA Spark clusters for local model training and inference, self-hosting AI models, n8n for workflow automation, and experimenting with CUDA-accelerated edge inferencing and local-first tooling. Always tinkering.


LinkedIn · Email · cascadeflow

Pinned Loading

  1. lemony-ai/cascadeflow lemony-ai/cascadeflow Public

    Cascading runtime for AI agents. Optimize cost, latency, quality, and policy decisions inside the agent loop.

    Python 305 97