Ameet S. Deshpande

Optimization, control algorithms and the protocols for energy networks.

I build optimization and control algorithms for physical energy systems — virtual power plants, home batteries, wind turbines, EV charging networks, and electric propulsion — and the protocols those systems trade over: the schemas, contracts and settlement rules that let independent parties exchange energy and flexibility and agree on what was owed. Mostly taking ideas from first principles through to production.

Most recently I designed the language primitives by which machines can express and agentically trade energy services, across several first-of-their-kind networks on the Beckn protocol — peer-to-peer energy trading, EV charging, and utility-led flexibility — including the trade ledger for India’s first interstate P2P energy network. Before that, at Tesla Energy on the Autobidder team, I was one of two engineers who built the trading algorithms behind the Texas virtual power plant that bids solar and storage from thousands of homes into day-ahead and real-time markets. Earlier, at GE Renewable Energy and GE Global Research, I worked on extreme-event controls, blade-twist recovery, load estimation and pitch-bearing prognostics for the global wind fleet.

I hold a Ph.D. in Mechanical and Aerospace Engineering from UC San Diego, where I worked on curse-of-dimensionality-free methods for solving the Hamilton–Jacobi–Bellman PDE, and a B.Tech. from IIT Bombay. Thirteen patents.

Based in Folsom, California.

Elsewhere

Writing

Notes and longer-form thinking live under writeups.

Selected publications

A. S. Deshpande. Bridging a gap in applied Kalman filtering: estimating outputs when measurements are correlated with the process noise. IEEE Control Systems, 37(3):87–93, 2017.
A. S. Deshpande. Max-plus representation for the fundamental solution of the time-varying differential Riccati equation. Automatica, 47(8):1667–1676, 2011.
W. M. McEneaney and A. S. Deshpande. Payoff suboptimality and errors in value induced by approximation of the Hamiltonian. SIAM Journal on Control and Optimization, 51(5):3993–4015, 2013.

A full list of publications and patents is in the resume.