Email After Reading Post

how would you email this person after they had read this: I’ve been thinking a lot about how to explain what I actually do. I’m not a traditional engineer. I didn’t come up through clean systems or structured paths. Everything I know was built under pressure—learning fast, adapting faster, rebuilding myself over and over again. I work with local AI systems. Not APIs. Not black boxes. I build and run the full stack myself—models, orchestration, interfaces—because I care about ownership. About understanding the system end to end. About not outsourcing my thinking. Most of my work sits in this space: Local LLM orchestration (Ollama, llama.cpp) Agent-based systems using tools like LangChain and NetworkX Full-stack apps with Python (FastAPI/Django) and React/Next.js Rapid prototyping using what people call “vibe coding,” but done with structured context and intent I’ve spent years doing data annotation and RLHF work across companies like Meta, Scale AI, and Apple, which gave me a deep understanding of how these systems are actually trained—not just used. What makes me different isn’t just the tools. It’s how I think. I build fast. I learn fast. I go deep on problems most people avoid. I care about psychological models, personas, and how humans interact with systems—not just the code itself. Everything I’m working toward is centered on one idea: privacy-first, locally controlled AI that actually belongs to the person using it. If you’re working on anything in that space—or need someone who can take an idea from nothing to a working system quickly—I’m open to connecting. --- the person: linkedin.com/in/jeroen-veen

:::writing{variant="email" id="74829" subject="Your thoughts on local, privacy-first AI?"} Hi Jeroen, I appreciate you taking the time to read through my post. I know it’s a bit unconventional, but that’s also the point—I’ve built my approach outside of the usual paths, and it’s shaped how I think about systems and ownership. I’m particularly interested in connecting with people who are thinking seriously about where AI infrastructure is going—especially around privacy, local control, and reducing dependence on black-box APIs. From your background, it looks like you’ve been operating in spaces where those questions actually matter in practice, not just theory. I’d be curious to hear how you’re thinking about that tradeoff right now—centralized capability vs. decentralized control. If it’s relevant, I’m currently building and experimenting with: - Fully local LLM stacks (Ollama / llama.cpp) - Agent-based workflows with explicit reasoning graphs - Rapid prototypes that go from idea → working system in days, not months No pressure on anything formal—I’m more interested in a real conversation than a pitch. If you’re open to it, I’d enjoy hearing what you’re working on and where you see things heading. Best, Daniel :::