Hi there, I'm Ayush Singh - Know more👋
I build AI systems that turn complex technical problems into practical products.
My work sits at the intersection of Agentic AI, LLM systems, RAG, backend engineering, and applied machine learning. I enjoy taking ideas from experimentation to production - designing the workflow, connecting models and tools, building scalable APIs, evaluating system quality, and making the final product reliable in the real world.
Lately, I've been working on
- Multi-agent AI systems and tool orchestration using MCP
- Advanced RAG pipelines for better retrieval and grounded generation
- Agentic workflows with memory, tool calling, and multi-step reasoning
- LLM and RAG evaluation using frameworks such as RAGAS
- Observability and tracing for AI systems
- Production APIs with Python and FastAPI
- Testing, reliability, and quality monitoring for LLM applications
- Exploring better ways to connect AI models with real-world tools and data
I care less about building isolated AI demos and more about building systems that can be tested, observed, evaluated, and trusted.
If you're working on something interesting in AI, feel free to reach out.
