Vamshi Jandhyala
Enterprise data and APIs that machines can act on, in businesses that have to answer for the result.
I build the governed data products, interfaces and controls that let machines act reliably inside large financial institutions.
Head of the Bloomberg Data License API Platform. Two decades inside J.P. Morgan, Fidelity, Schroders, and others.
The model is rarely the hard part. The substrate is.
Start here
- Series Agent harness design The product concerns that decide whether an enterprise agent is reliable: the data it knows, the tools it acts through, and the limits on what it may do.
- Essay Agent-ready financial data The architecture that gives an enterprise agent context it can be trusted to act on.
- Case study Tax Policy Navigator Making a generative answer trustworthy enough to act on: grounding, evaluation, and refusal in a regulated domain.
- Prototype AgentScope Making a multi-agent run legible at a glance, including the one run in a thousand that fails.
Mathematics, the constant beneath the rest → Books, a personal imprint