← Energy board · written 2026-10-15 by claude-opus-5-5 from the pipeline's evidence only

evcc-io: open-source EV charge controller and home energy manager (Germany). Skoor 50, n=4, confidence 0.8

Data problem: the hiring signal, weighted 20 of the 50 points, comes from evcc.com, not evcc.io [source]. The sampled roles are emergency veterinary jobs in Michigan and Indiana [source]. Inference: this is a different organization, and the Skoor is overstated by 20 points.

What they built

evcc is an extensible EV charge controller and home energy management system [source]. Its stated goal is local energy management without relying on cloud services [source]. It supports many EV charger brands, EEBus, OCPP, smart switches, heat pumps and electric heaters [source]. The main repo is written in Go [source].

Record

Measured (signals, n=4):

  • 938 commits in 90 days across 1 energy repo [source].
  • 17 distinct committers in 90 days [source].
  • Website at evcc.io [source].
  • 46 open roles, but these are misattributed (see above) [source].
  • From the GitHub metadata:

  • The repo was created 2019-12-06 and last pushed 2026-10-14 [source].
  • It has 7,368 stars and 1,557 forks [source].
  • The org was created 2021-03-25, is based in Germany, is verified, and has 21 public repos [source].
  • The project was featured in PV Magazine and c't Magazin [source].
  • Inference: the commit pace and fork count point to an active open-source project. They do not show a company. Founders, legal entity, funding, revenue and customers are not in the evidence.

    Why now

    Market timing is not in the evidence. Inference: a German, local-first controller that works with many hardware brands fits households pairing solar with EVs and heat pumps. No demand data supports this.

    Unknowns

    Questions for the room

    What would move the Skoor

    It would rise with evidence of a legal entity and named technical founders, a verified install count, or paying OEM or installer customers. It would fall if the misattributed hiring signal is removed (Inference: to about 30), or if commits turn out to come from one or two volunteers.