Forecast Engineer (@ForecastEng) header
Forecast Engineer (@ForecastEng) profile picture

Build forecasting systems, not just forecasts. ForecastOps, engineering, backtesting, monitoring, and business reporting. Register for my next workshop.

Pittsburgh, PA
Joined June 2026
Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)Photo by Forecast Engineer (@ForecastEng)
Python has powerful time series libraries. Case in point: skforecast Let me explain:
2
12
2
88
7,954
Knowing ARIMA or Prophet is not enough anymore. Companies need people who can build forecasting systems. That's the shift from Data Scientist to Forecast Scientist. Wednesday, July 8, I'll show my system live in Python. Join Here (670 registered): learn.business-science.io/fo…
2
1
12
1,219
Forecast Engineer retweeted
The gap in most forecasting education: You learn the model. But not the system. On July 8, I'm showing the full workflow live: Data → Forecast → Backtest → Production-ready system Free. Python. Full code. 670+ already registered. Save your seat → learn.business-science.io/fo…
2
5
884
Forecast Engineer retweeted
Most courses train you on clean data. Then you hit real supply-chain data full of zeros — and everything breaks. Intermittent demand is one of the biggest forecasting problems nobody teaches. Let's unpack it. Bookmark this for later. 🧵
2
3
1
9
871
The gap in most forecasting education: You learn the model. But not the system. On July 8, I'm showing the full workflow live: Data → Forecast → Backtest → Production-ready system Free. Python. Full code. 670+ already registered. Save your seat → learn.business-science.io/fo…
2
5
884
Most courses train you on clean data. Then you hit real supply-chain data full of zeros — and everything breaks. Intermittent demand is one of the biggest forecasting problems nobody teaches. Let's unpack it. Bookmark this for later. 🧵
2
3
1
9
871
Companies don't want another ARIMA model. They want engineers who can build forecasting systems that actually ship to production. Wednesday, July 8th — I'm sharing mine live in Python. Free. Full code. 670+ already registered. Save your seat → learn.business-science.io/fo… 
1
46
That's a wrap! If you enjoyed this thread: 1. Follow me @ForecastEng for more of these 2. RT the tweet below to share this thread with your audience
Most courses train you on clean data. Then you hit real supply-chain data full of zeros — and everything breaks. Intermittent demand is one of the biggest forecasting problems nobody teaches. Let's unpack it. Bookmark this for later. 🧵
37
90% of data science programs never teach you time series. So most data scientists never learn the one skill that makes them hard to replace. July 8, I'm closing that gap — live, free, full code. 670+ already in. Register Here: learn.business-science.io/fo…
1
2
14
1,322
There are now 8+ time-series foundation models. Here's which ones are actually worth trying. 🧵
1
7
1
20
1,662
Wednesday, July 8th — I'm sharing mine live in Python. Free. Full code. 670+ already registered. Save your seat → learn.business-science.io/fo…
1
3
45
That's a wrap! If you enjoyed this thread: 1. Follow me @ForecastEng for more of these 2. RT the tweet below to share this thread with your audience
There are now 8+ time-series foundation models. Here's which ones are actually worth trying. 🧵
1
52