Industries
RunnerTau's reasoning starts wherever the time-series data is richest and the failure modes are best understood — then extends to adjacent domains that share the same underlying physics or the same reasoning structure.
Energy
Energy is where RunnerTau is proven. RunnerTau 7B and 4B already reason about energy problems the way a domain expert would — understanding battery systems well enough to explain charge behavior, flag abnormal degradation, and reason about dispatch decisions against real operating constraints.
This is also where our consultancy engagements start: assessing whether an energy operator's telemetry, documentation, and asset context are reasoning-ready, then building the TEMPOral knowledge layer that lets RunnerTau reason over their specific assets.
Home battery
"Optimize my 20kWh home battery schedule for tomorrow, using 72h history and tomorrow's pricing to minimize cost."
Grid battery array
"Can the fleet commit to the 08:00 frequency response window given current SoC across sites?"
Wind turbine
"Why did output drop 15% this morning?"
BTS node / backup power
"Can we forecast tomorrow's load despite a failing power meter?"
Data centres
Data centres share the same underlying physics as energy assets: power draw, thermal limits, and capacity constraints that telemetry alone doesn't explain. As compute density rises, the questions operators need answered — can we admit this workload, is this rack's telemetry trustworthy, will cooling hold — are time-series reasoning problems.
We're extending RunnerTau's energy-domain reasoning into this vertical next, building on the same asset-grounding approach already proven with battery systems. It is also where security and operations meet: an intrusion can first look like a power or cooling anomaly, and the evidence is spread across the same logs and telemetry.
GPU cluster
"Can we admit tomorrow's training workload without exceeding cooling limits?"
AI compute campus
"Rack 16's power telemetry is degraded. Can we still plan capacity confidently?"
Security & Defense
Security and defense operations generate two kinds of evidence that rarely meet: text that arrives over time — system logs, alerts, incident tickets, maintenance notes, field reports — and numeric telemetry such as network flows, sensor feeds, power and thermal signals. Detectors flag that something deviated. Working out why, what is likely to happen next, and what to do about it is left to an analyst who joins the two by hand.
These are the same reasoning problems we solve in energy: root-cause analysis, forecasting grounded in real constraints, and a decision at the end. What changes is that the deciding evidence is often a line of text that a signal-only model can't see. We're extending RunnerTau's asset-grounded reasoning into this vertical next. The same reasoning can be adapted to your own logs, telemetry, and asset context — the examples below are a starting point, not a boundary.
Security operations
"Failed logins spiked on the VPN gateway at 02:10, and one host's outbound traffic doubled an hour later. Are these related — and is it exfiltration or the scheduled backup?"
Industrial control systems
"Valve-position telemetry at the pump station drifted after last night's firmware change in the maintenance notes. Fault, misconfiguration, or tampering?"
Sensor network
"Three field sensors stopped reporting within minutes of each other, and the comms log shows a relay reboot. Equipment failure, interference, or just the relay?"
Facility perimeter
"Fence vibration alarms fired at three sites while badge readers logged errors. Is this coordinated probing, or the storm front?"
Automotive
Automotive powertrains and battery packs are, structurally, the same reasoning problem as stationary battery systems — charge and discharge behavior, degradation, and thermal limits — with the added complexity of mobility and fleet-level telemetry.
Having proven RunnerTau's reasoning on stationary battery systems in energy, automotive is a natural extension of the same underlying model family.
Vehicle battery pack
"Diagnose the cell imbalance from the uploaded charge/discharge readings."
Fleet telemetry
"Which vehicles in the fleet are showing early signs of battery degradation?"
We're expanding RunnerTau's domain coverage deliberately, one vertical at a time. Tell us what you're working on.
Talk to us →