Live from ICC 2025 in Sacramento: Greg Sloyer (Industry Principal, Manufacturing at Snowflake) joins the Energy Tech Podcast to break down how OT data from Ignition can flow—with context—into Snowflake using IoT Bridge, MQTT/Sparkplug, and Snowpipe Streaming, then combine with ERP/CRM data for real business outcomes. We dig into Cortex (semantic layer, NLQ, search over manuals/service records), DataOps/DevOps realities, and pragmatic analytics for energy: compression fleets, maintenance, and production trends.

Guest

Greg Sloyer, Industry Principal—Manufacturing, Snowflake.

What you’ll learn

- Why ICC 2025 is a “level up”: new venue, Ignition 8.3, PROVE IT

- Edge → Cloud: UDT context, MQTT/Sparkplug, IoT Bridge & Snowpipe Streaming

- OT + IT together: time series + ERP/CRM for bigger, multi-site problems

- PoC fast paths for regional operators (free trials; zero-code landing)

- Democratized analytics: run anomaly detection via SQL

- Cortex: semantic layer, natural-language questions, and search over manuals & service history

- Energy-specific cadence: hourly/daily/7-day moving averages vs sub-ms myths

- Fleet view of compressors: MTTR/MTBF, asset health, and market-aware maintenance

- When to use Snowpark/Streamlit vs NLQ for quick operator answers

Links & resources

• Snowflake for Manufacturing & Energy: https://www.snowflake.com

• IoT Bridge (with Cirrus Link): https://cirrus-link.com/

• Ignition 8.3 (Inductive Automation): https://inductiveautomation.com/news/inductive-automation-releases-ignition-83

• Opsite Energy – Insight-Driven Remote Operations: www.opsiteenergy.com

00:10 Welcome to Energy Tech Podcast @ ICC 2025

00:26 Guest intro: Greg Slayer (Snowflake)

00:49 Greg’s role: Industry Principal, Manufacturing

01:44 ICC “level up”: new venue & more room for real OT/IT talks

02:46 Why ICC 2025 matters: Ignition 8.3 + PROVE IT track

03:50 Where Snowflake fits with Ignition & IA

04:21 Getting shop-floor OT data into Snowflake (since 2023)

05:03 IoT Bridge + MQTT/Sparkplug → OT+IT data together

06:01 Edge-first: UDT context, Snowpipe Streaming, zero-code landing

07:24 Context that business users understand (units, semantics)

08:33 Add sensors/tags → auto-populate in Snowflake for data science

09:05 Fast start for regional energy players (trials & PoCs)

10:51 Democratized analytics: SQL anomaly detection on time series

11:51 Modeling options, right cadence (hourly/shift vs sub-ms)

12:42 Snowflake Cortex: semantic layer, NLQ, medallion patterns

13:28 Energy cadence: 7-day moving averages, hourly/daily ops

15:44 Maintenance reality: dispatch latency & practical “near real-time”

16:28 Cortex Analyst & Search: manuals, service records, tech assist

19:16 Compression fleet use case: asset health + market signals

20:26 Fleet-level KPIs: MTTR/MTBF across sites & vendors

21:01 Snowpark & Streamlit vs NLQ answers for ops

22:32 “Advanced analytics,” not hype: optimization + ML in Snowpark

24:26 Wrap-up, resources & how to try it; Like & Subscribe

Music: Uygar Duzgun / “Fast Life” / courtesy of www.epidemicsound.com