"As far as the laws of mathematics refer to reality, they are not certain
and as far as they are certain, they do not refer to reality"
Hi, I'm Aastha.
I build AI systems that hold up in production — not in demos. My work sits at the intersection of agentic AI, context engineering, and system architecture. From orchestrating stateful multi-agent LangGraph pipelines to shipping Graph-RAG systems at scale, I'm drawn to the problems that don't have clean answers yet — where the gap between a model that scores well on benchmarks and one that works reliably under real traffic is still wide open.
I've shipped HR automation agents, legal AI platforms, and enterprise RAG systems, owning architecture end to end. I believe in the silent power of well-designed systems — where every decision upstream echoes in how gracefully things fail downstream.
Let's build something that matters.