Episode 15

Taking HEMS from startup to scale-up

Episode 15
·
50 mins
·
December 17, 2024

Taking HEMS from startup to scale-up

Georgia talks to early gridX teammates Baptiste Feron (Head of EMS) and Robert Fritzsche (Software Engineer) about building and scaling a HEMS: early challenges, real-world device quirks, team evolution, resilient architecture and practical tips for startups entering smart energy.
Listen on:

What you’ll learn in this episode

From train tables to test labs: The HEMS origin story

In the early days, the team did everything – coding, installs, support – often literally on the move. Over time, gridX shifted from end-customer experiments to a platform powering partners at scale. The mission stayed steady: connect DERs and orchestrate them intelligently.

Control in the wild: why stability beats elegance

Real devices don’t follow textbooks. Batteries respond at different speeds, EVs can be finicky (hello, harmonics) and firmware variants change behavior. The cure: event-driven control, guardrails against oscillation and algorithms tuned for delay, jitter, and partial failure, so charging doesn’t stutter and homes stay smooth.

Interoperability at scale: Same logic, many personalities

A battery is a battery – until it isn’t. gridX abstracts OEM quirks behind clear interfaces and a unified data model, so one EMS can steer hundreds of device types. Strong observability, alerts, canary releases and instant rollbacks keep weekly/monthly updates safe across thousands of unique setups.

From two teams to many: shipping fast without chaos

As scope grew (new markets, use cases, regulations), gridX split monoliths into focused, cross-functional teams (engineers + EMS experts + product). Clean interfaces, independent deploys and shared language (Go across stack) keep cognitive load low – and velocity high.

Startup playbook: What they wish they knew

  • Write the problem first. Precise problem statements reveal half the solution.
  • Test on real hardware early. Simulations lie; field data tells the truth.
  • Abstract aggressively. Stable interfaces outlive devices and firmware.
  • Be resilient by default. Expect timeouts, jitter, partial control, and fallbacks.
  • Narrow the first scope. Win with a few setups, then expand.
  • Ship safely. Feature flags, staged rollouts, observability, quick rollback.

Key quotes: Interfaces are everything

“Real hardware never behaves as expected. Your EMS has to.” — Baptiste Feron

“You shouldn’t need to read the code to understand the system – read the interface.” — Baptiste Feron

“The faster you react to new measurements, the more potential you unlock.” — Robert Fritzsche

“Be close to OEMs and installers. You’re shipping one solution, together.” — Robert Fritzsche

Receive new episodes in your inbox
Tune in bi-weekly as we shine a spotlight on the different facets of smart energy technology.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.