On the website of a year-old New York startup called Strobe Power, the pitch fits on a single line: "One controller, every asset, real-time dispatch. Cut your energy bill 20-40%+" [Strobe Power, 2024]. That sentence is doing a lot of work. It is asking a factory manager in Ohio or a cold storage operator in Texas to hand over the orchestration of their boilers, batteries, rooftop solar, backup generators, and HVAC to one piece of software, and to trust that the software will save them real money without dropping a load at the wrong moment.
It is the right pitch for the right year. The U.S. commercial and industrial power buyer is dealing with two trends at once: rates climbing in most ISOs, and reliability quietly getting worse as transformers age and interconnection queues stretch into the 2030s. Strobe is positioning itself as the autonomous operator that sits on top of whatever mix of assets a site already has, and squeezes the bill down through load shifting, peak shaving, and dispatching the cheapest electron available at any given fifteen-minute interval [f4.fund].
The bet
Strobe's wedge, as described on its own site and in a writeup from MIT's Orbit program, is the controller layer [MIT Orbit, 2024]. Most C&I sites today run a patchwork: a building management system from one vendor, a battery EMS from another, a generator controller from a third, and a utility tariff spreadsheet maintained by whoever has been there longest. Strobe's argument is that the savings live in the seams. If one piece of software can see the tariff, the weather, the on-site generation, the storage state of charge, and the demand charge ratchet all at once, it can make dispatch decisions that no single-vendor system can.
That 20-40% bill reduction claim [Strobe Power, 2024] is at the high end of what optimization software typically promises in this category, and it implicitly assumes a site with enough flexible load and on-site assets to actually move the needle. A pure office building with no battery and no generator is not the customer. A food processor with refrigeration, a backup genset, and a demand charge that accounts for a third of the bill is exactly the customer.
Why it could be big
The addressable wedge here is genuinely large. C&I electricity spend in the U.S. runs into the hundreds of billions annually, and the share of sites adding batteries, solar, or backup generation is climbing as outage costs rise and the Inflation Reduction Act's investment tax credits remain in force. Every one of those new assets is a candidate for an orchestration layer. The MIT Orbit framing of Strobe is exactly that: a response to "the alarming trend of rising power outages and the aging U.S. grid" for C&I businesses [MIT Orbit, 2024].
The company is also showing up in the right rooms. It is listed in the climate and clean energy portfolio of f4.fund, an early-stage program that backs technical founders in energy and industrial software [f4.fund]. The MIT Orbit Launchpad listing suggests at least some connective tissue to the Cambridge research and founder community [MIT Orbit, 2024]. Neither is a Series A validation, but both are signals that technically literate allocators have looked at the controller-layer thesis and decided it is worth a conversation.
Here is the back of envelope. A mid-size U.S. food processing plant might spend roughly $2M a year on electricity (estimated). A 25% reduction, the midpoint of Strobe's claimed range, is $500K. If Strobe charges a typical optimization-software take rate of 15-20% of savings, that is $75K to $100K of annual contract value per site (estimated). Land 200 such sites and you are at a $15M to $20M ARR business without leaving the Midwest. The category math works. The execution math, which is whether a pre-seed team can actually integrate with the long tail of BMS, inverter, and genset protocols at each site, is the harder question.
Team and traction
Strobe was founded in 2024 and is based in New York [PitchBook]. Public disclosures on the founding team, funding round, and customer count are not yet on the record, and the company is at the stage where most of what exists is the product thesis and the early controller software itself. What is on the record is consistent: the website, the MIT Orbit page, and the f4.fund listing all describe the same product in the same words, which at least suggests a team with a clear and stable articulation of what they are building [Strobe Power, 2024] [MIT Orbit, 2024] [f4.fund].
| Strobe Power at a glance | |
|---|---|
| Founded | 2024 [PitchBook] |
| Headquarters | New York, NY |
| Stage | Pre-seed |
| Claimed bill reduction | 20-40%+ [Strobe Power, 2024] |
| Named competitors | GridStrong, Lumora |
The honest counterfactual
The bear case is the integration tax. Companies have been trying to build the universal C&I energy controller for at least fifteen years, and the graveyard includes well-funded names. Each site is a custom integration project, sales cycles run six to twelve months, and the incumbent building management vendors (Johnson Controls, Siemens, Schneider) have deep relationships and the ability to bundle. Named competitors GridStrong and Lumora are chasing the same wedge with similar pitches. The bull answer, and it is the one Strobe's own materials lean into, is that the asset mix at C&I sites has changed enough in the last three years (more batteries, more on-site solar, more EV chargers, more interval data) that an autonomous-dispatch-first product can finally outrun the legacy BMS approach, which was designed for a world where the only flexible asset was the thermostat [Strobe Power, 2024].
What to watch
The next twelve months for Strobe come down to two things: a named lighthouse customer with a verifiable savings figure, and a priced seed round that puts a real investor name on the cap table. Either would meaningfully de-risk the story. A published case study from a recognizable industrial brand, even a single site, would do more for Strobe's sales motion than another round of website copy. Watch f4.fund's portfolio updates and the MIT Orbit cohort announcements for the first hard signal.
The company Strobe most clearly has to beat is Schneider Electric's EcoStruxure, the incumbent that already sits inside a large share of U.S. industrial sites and that any factory manager will mention in the first meeting. Out-orchestrating EcoStruxure on a per-site basis is a tall order. Doing it at a tenth of the integration cost is the only version of this story that ends with Strobe as a category winner.