LinkedIn·Friday, 14 August 2026·12d ago
Time-series foundation models promise anomaly detection with no feature engineering. We tested the promise. The pitch is appealing: point a…
Dataminded
4,964 followers
Time-series foundation models promise anomaly detection with no feature engineering. We tested the promise.
The pitch is appealing: point a pretrained model at a time series, skip the feature work, and get anomalies. During our Innovation Days, Ketan Jaltare and his team tested MOMENT-1-large on the full Numenta Anomaly Benchmark and compared both MOMENT-1-large and Chronos-2 on representative NAB series.
The verdict: promising, but not production-ready yet.
The interesting part is understanding where these models fall short today, and what it would take to close the gap.
Interested in exploring how foundation models can be applied to your data challenges? We run free knowledge-sharing sessions for data teams, at your office, on the topic of your choice.
Book a session with Jonny Daenen: https://lnkd.in/ecEHzhxu
Cross-referenced
Related on the wire
Last week we hosted our Agentic AI Skills for Data Practitioners workshop at our Antwerp office, and what an evening it was! 🙌 We were…
Many organisations started a Data Mesh journey. Far fewer reached the outcomes they hoped for. Why? In the latest episode of The Data…
Most of what you ask AI to code isn't that hard. So why run it through your most expensive model? Teams often route routine work (Terraform…
Think your agent setup can hold up under real conditions? Prove it. Same model. Same tasks. Your harness, your context. Dataminded's data…
Most AI agent demos never touch a real pipeline. Ours does, and it gets scored on it. On 26 August in Antwerp, we're running a hackathon…
An AI agent scored 37 out of 100 on questions about our own data. The same model later scored 100 out of 100. Ask an AI agent a question…