The spirits track: machine learning across distillation and the long, slow business of maturation — cut-point selection, new-make spirit character, the angel’s share, cask inventory, congener development, blending consistency, and bottling maturity. Written for distillers, with the same honest line on what data can and cannot tell you across years in the warehouse. The full track, in order:
-
A Cask Maturation and Angel's-Share Dashboard in Tableau
Build a Tableau cask maturation dashboard tracking age, strength and angel's-share evaporation loss, with projected ready dates and AI-driven monitoring.
-
A Rackhouse Microclimate Dashboard in Tableau
Build a Tableau rackhouse microclimate dashboard mapping warehouse temperature and humidity by position to maturation loss, with AI outlier explanations.
-
A Maturing-Stock Inventory and Valuation Dashboard in Tableau
Build a Tableau maturing-stock dashboard valuing whisky inventory by age, cask and strength, tracking tied-up capital and casks reaching age targets.
-
A Whisky Blending and Sensory Dashboard in Tableau
Build a Tableau blending dashboard combining cask sensory and GC profiles, candidate-vatting what-if analysis and consistency checks against house style.
-
A Distillery Production and Spirit-Yield Dashboard in Tableau
Build a Tableau distillery dashboard tracking litres of pure alcohol per tonne, yield by run, cut performance and energy per LPA, with AI monitoring.
-
Can AI Pick Distillation Cut Points (Heads, Hearts, Tails)?
Can AI choose distillation cut points? How models use ABV, temperature, time and congener data to make heads, hearts and tails cuts more consistent.
-
Whiskey Tasting, Captured and Analysed: Power Apps, Power BI, and AI
Capture cask-sample and vatting scores in Power Apps linked to ERP, track maturation in Power BI, and use AI to cluster casks and flag off-notes — nose still decides.
-
Predicting New-Make Spirit Flavour
How AI and data science model new-make spirit congeners from ferment, still shape and cut, so distillers steer flavour before the cask.
-
Forecasting the Angel's Share: Modelling Cask Evaporation Loss
How AI and data science forecast the angel's share, the ~2%/year cask evaporation loss, to plan fill volumes and value maturing whisky stock.
-
AI for Cask Selection and Maturing-Stock Inventory
How AI forecasts which casks hit a flavour and age target and when, to optimise tied-up maturing-stock capital and pick casks for whisky releases.
-
Predicting Congener Evolution During Maturation
How AI models congener and wood-extractive evolution over years from cask type, char and microclimate, cutting how often you must sample whisky.
-
Optimising Rackhouse Microclimate With AI
How AI and warehouse sensors map rackhouse microclimate to reduce cask-to-cask variation in extraction and angel's share, and guide smarter cask placement.
-
AI for Whisky Blending and Vatting Consistency
How AI optimises blend recipes — cask proportions, ages and wood types — to hit a consistent house profile batch after batch at the lowest cost of premium stock.
-
Predicting Optimal Bottling Time and Maturity
How AI models when a cask hits its age, flavour and ABV target so you avoid over-maturing or under-maturing — while sensory judgement stays the final call.
-
AI for Gin Botanical Extraction and Recipe Consistency
How AI models maceration, vapour infusion and the cut, and compensates for botanical lot variability to keep a gin recipe consistent batch after batch.
-
AI for Distillery Energy and Spirit Yield
How AI optimises spirit yield — litres of pure alcohol per tonne — and cuts energy across fermentation, distillation and the cut, with honest limits on the yield-flavour trade-off.
-
Databricks for Distilleries: 20 Use Cases
How a distillery uses Databricks end to end — ingestion, real-time monitoring, the Delta Lakehouse & Spark, BI and AI — across 20 concrete use cases grouped by capability.
-
How a Distillery Should Start with AI and Gen AI: The Phases
A phased roadmap for a distillery adopting AI and generative AI — from collecting data to dashboards, predictive models, GenAI copilots and agents — with what to do, need and watch at each phase.
-
Databricks Across the Distillery Business, Vertical by Vertical
A department-by-department tour of where Databricks helps a distillery — from the floor through quality, supply chain, sales, marketing, finance and compliance — on one governed platform.
-
Claude AI and Claude Code for Distilleries: Where the Anthropic Ecosystem Helps
Where Claude, Claude Code, the API, agents and MCP help a whiskey distillery — new-make R&D, distillation, cask and warehouse, sales, marketing, excise compliance and knowledge — and where a human must stay in the loop.
-
Snowflake for Distilleries: 20 Use Cases
How a distillery uses Snowflake end to end — ingestion, real-time monitoring, Dynamic Tables & Snowpark, BI and AI — across 20 concrete use cases grouped by capability.
-
Snowflake Across the Distillery Business, Vertical by Vertical
A department-by-department tour of where Snowflake helps a distillery — from the floor through quality, supply chain, sales, marketing, finance and compliance — on one governed platform.
-
Microsoft Fabric for Distilleries: 20 Use Cases for Whiskey (and 3 Case Studies)
How a whiskey distillery uses Microsoft Fabric — OneLake, Real-Time Intelligence, Lakehouse, Direct Lake and Copilot — across 20 use cases from still telemetry to cask maturation and maturing-stock valuation, plus three case studies.
-
IoT in the Distillery: Sensors from Mash to Cask
An IBD-grounded guide to IoT in distilling — sensors at mashing, washback fermentation, the wash and spirit stills, the spirit cut, and the maturation warehouse, plus the edge-to-cloud stack and AI.
-
What Is AI, Really? A Distiller's Plain-Language Guide
No jargon, no hype: what 'AI' actually means for a distillery, why the word covers three very different things, and how to tell a useful tool from a buzzword before you spend a rupee or a dollar on it.
-
What Is Machine Learning? Pattern-Finding for the Distillery
Machine learning explained for distillers without the maths: software that learns patterns from your own cask and process records, then predicts the next case — angel's share, cut points, bottling maturity — plus exactly where it quietly breaks.
-
What Is Generative AI? The Difference That Matters for Distillers
Generative AI explained for distillers: the chatbot kind of AI that writes, drafts and explains in plain language — what it's genuinely good at in a distillery, why it states wrong cask numbers as confidently as right ones, and how grounding fixes that.
-
AI vs Machine Learning vs Generative AI: Which Is Which in the Distillery
A distiller's decision guide to the three terms everyone confuses — what each is, what data it needs, how it fails, and a simple test for matching a real distillery problem to the right one instead of buying the buzzword.
-
Where a Distillery Actually Starts with AI
The capstone of the foundations series: not with a tool, but with one measured record. A distiller's order of operations — data, then dashboards, then machine learning, then generative AI — and the honest reason most AI projects fail before they begin.
-
Descriptive, Diagnostic, Predictive, Prescriptive: The Four Analytics in a Distillery
The four classic types of data analytics translated into distillery language — spirit yield, cut points, the angel's share, cask allocation — and where Microsoft Fabric's lakehouse, real-time intelligence and Direct Lake actually fit each one.
-
Why Power BI Dashboards Die in Distilleries and Breweries: The Domain Gap
A skilled Power BI developer walks into a distillery and eighteen months later the dashboards are abandoned. Not a tooling failure — a translation failure: units, batch genealogy, process time and KPIs that don't match how distillers and brewers actually think.
-
The Translator: When the Brewer Is Also the Power BI Developer
What changes when one person has stood at the mash tun, the spirit safe and the crush pad — and also holds the Power BI and Microsoft Fabric certifications. Why the hybrid profile builds production dashboards that survive, and the honest limits of being one.