From Redundancy to Wine Tech: The Story Behind SavvySipper

There are several ways to respond to being made redundant.

You can spend six months furiously updating LinkedIn.

You can convince yourself that drinking coffee whilst scrolling through job adverts counts as a productive morning.

You can become extremely interested in the phrase "open to opportunities", despite having absolutely no idea what those opportunities might actually be.

Actually... don't get me wrong. I did all those things until I was brought in to consult for a leading Cruise line to build a new B2B2C division.... during my time there is when the idea for SavvySipper came to me.

So... despite all of the frustrations, the low times, the lonely times, the "Sorry but we've gone with the internal candidate" or "The role is on hold" or worst of all, the consistent ghosting from recruiters who never even had the human decency to get back with any feedback... Apparently, you can decide to build an AI-powered wine technology tool from scratch despite having no formal background in software development.

Welcome to the rather unlikely story of SavvySipper.

The Bit Where Everything Changed

At the beginning of 2025, I was made redundant.

After having spent years building a career in travel, following several years in Entertainment, it was not exactly the plot twist I had envisioned. Until a new CEO arrived at my former company, things had been going very well professionally, and I had plenty of experience, responsibility and momentum behind me.

Then suddenly, the future looked rather less certain.

I won't bore you with the corporate details, partly because I had a six-month non-compete, partly because there are always two-sides to the story (sometimes more), partly because there are such things as rose tineted glasses, and partly because nobody needs another overweight middle-aged bloke on the internet writing a thinly disguised corporate memoir.

But redundancy has a funny way of giving you an unwanted amount of time to think.

And throughout 2025, one thought kept coming back.

What if I built my own thing?

Not another job.

Not another corporate ladder.

Something I owned.

Something I could learn from.

Something where I could wake up and genuinely have a reason to figure out something I didn't know how to do the day before.

That last bit turned out to be rather important.

The Idea Wasn't Originally "Let's Build a Wine Company"

The seed for SavvySipper came later, during a business meeting (aka a regular wine fuelled get-together with a fun client now working with me at the Cruise line) at Quill Wine Bar in London Bridge in early September 2025.

As I was waiting for her arrival I started looking properly at the wine list.

One thing in particular caught my attention.

The pricing of the different serving sizes simply didn't make sense.

It was significantly better value to order 500ml carafes than the 750ml bottles.

Why?

That question bothered me considerably more than it probably should have.

Most people would simply order the bottle, order the carafe or ask the person next to them what they fancied.

I started doing maths.

Because apparently this is what happens when you have too much time on your hands.

The more I looked, the more interesting it became. Restaurant wine pricing isn't simply about the cost of the bottle. There are different serving sizes, glass prices, bottle prices, retail benchmarks, perceived value and psychological pricing decisions all sitting underneath that innocent-looking wine list.

And suddenly I had a question.

What if a normal person could scan a restaurant wine list and actually understand what they were looking at?

Not after spending three hours learning wine terminology.

Not after subscribing to some extraordinarily serious wine service.

Not after asking a sommelier whether a £42 bottle represents "good value", only to receive an answer involving gravelly soils and an uncle in Burgundy.

What if technology could simply do the work?

That was the lightbulb moment.

From Wine List to Wine Intelligence

The original concept was deliberately simple.

Take a restaurant wine list.

Read it.

Identify the wines.

Look at the serving sizes.

Calculate the margins.

Find the best Food Pairings.

Add some Sommelier Insights and Trivia about the venue, the wines or vineyards etc

Show the approximate Vivino Score - NB: They have no public API by the way

Compare them against real-world retail pricing.

Then turn all of that rather tedious information into something a normal human being could actually understand.

That became the foundation of SavvySipper.

The idea was never to tell people what wine they should drink.

Quite the opposite.

If you want to spend £200 on a bottle of Bordeaux, crack on!

The idea was to give people enough information to make their own decision.

Drink what you like.

Pay what you think is fair.

And, ideally, avoid being taken for an absolute mug.

Then I Had to Actually Build the Thing

This was where the story became considerably more interesting.

Because I could understand the business idea.

I could understand the wine problem.

I could understand the customer.

What I couldn't do was build a technology platform.

I had to learn.

And learn.

And then learn why the thing I had just learned had broken something I had built two hours earlier.

Over the following months, I went down a rather spectacular technology rabbit hole.

I learned coding.

I learned how websites actually work rather than simply assuming that someone in "IT" deals with it.

I learned GitHub and version control.

I learned how to deploy applications using Vercel.

I learned how to work with databases using Supabase.

I learned how to manage repositories and development workflows with GitHub.

I learned how to build email infrastructure with Resend.

And then AI-assisted development arrived and changed the rules again.

Tools such as Google Antigravity became part of the development process, giving me something that was somewhere between a coding assistant, technical sounding board and incredibly patient colleague who never appears to get bored of the same question being asked for the seventh time.

It was extraordinary.

It was also occasionally maddening.

The Development Chat History Should Probably Be Classified

If you could read through the development conversations from the last several months, you would probably assume I had either become a software engineer or suffered a nervous breakdown.

There were questions like:

"Why has this Vercel deployment suddenly stopped working?"

"Why is this database query returning the wrong wine?"

"Why has fixing this button broken three other buttons?"

"Why does the mobile version look like that?"

"Why is this CSS card grid behaving like a stroppy teenager?"

"Why has changing one field in the database somehow caused the entire application to develop an existential crisis?"

And, on particularly exciting days:

"What the hell have I just done?"

The joy of building something yourself is that every problem is yours.

The downside of building something yourself is that every problem is yours.

AI Didn't Build SavvySipper for Me

This is probably worth saying because there is a rather lazy narrative around AI-assisted development.

People hear that AI can write code and assume the machine simply produces an application while the human sits nearby drinking coffee.

That has not been my experience.

AI has dramatically accelerated what I can learn, test and attempt.

It has helped me understand concepts that would previously have taken weeks to get my head around.

It has helped me debug code.

It has helped me structure databases.

It has helped me understand deployment errors.

It has helped me explore ideas that would probably have remained ideas without it.

But it has also made me responsible for understanding what the bloody thing has actually done.

That distinction matters.

When something breaks, saying "the AI did it" isn't particularly useful.

The website still doesn't work.

The database still doesn't work.

And the wine scanner still isn't going to magically fix itself because you have stared at the screen angrily.

Trust me. I've tried.

From Prototype to Something Real

The first versions of SavvySipper were considerably more basic than what exists today.

The early development focused heavily on extracting information from wine lists using OCR and then trying to identify what the restaurant was actually selling.

That sounds straightforward.

It wasn't.

Restaurant wine lists are wonderfully inconsistent documents.

Different fonts.

Different layouts.

Different vintages.

Different bottle sizes.

Different spellings.

Different ways of describing exactly the same wine.

And, occasionally, a photograph taken at an angle that appears to have been captured during an earthquake.

The more I built, the more obvious it became that simply reading the menu wasn't enough.

SavvySipper needed to understand the wine behind the text.

That meant building data structures, matching producers, normalising wine names, comparing retail prices and creating a scoring system that could turn all of that information into something useful.

The Four Numbers That Became Five

The platform evolved into a broader wine intelligence system built around several core measurements.

Margin Fairness looks at the relationship between restaurant pricing and wider retail value.

Crowd Score brings consumer sentiment into the equation.

Glass Tax looks at the economics of ordering wine by the glass rather than the bottle.

Menu Discovery considers the breadth and diversity of the wine list.

And then there are the Sommelier Insights - the part designed to turn the numbers into something that actually sounds like useful wine advice rather than an accounting spreadsheet.

That combination is what makes SavvySipper different from simply asking whether a wine is expensive.

A wine list can be expensive but excellent.

It can be cheap and not cheerful.

It can be brilliantly curated but commercially chaotic.

It can have wonderful wines but absurd glass pricing.

The interesting bit is understanding the whole picture.

Why I Wanted SavvySipper to Be About More Than Wine

The deeper I got into the project, the more I realised this wasn't really a wine problem.

It was an information problem.

Restaurants have information.

Retailers have information.

Wholesalers have information.

Wine producers have information.

But the person sitting at the table often has remarkably little.

They have a list of wines.

They have some prices.

They might recognise a grape or two.

And they have their dining companions staring at them waiting for someone to make a decision.

That felt like an opportunity.

Give the consumer better information.

Give restaurants better insight into their own lists.

Give retailers a route from discovery to purchase.

Give wholesalers better intelligence about what venues actually need.

And give wine lovers a way of discovering bottles without needing to become a part-time Master Sommelier.

The Most Ridiculous Part? I'm Still Learning

This is probably my favourite part of the whole thing.

SavvySipper isn't finished.

I'm not sure it ever will be.

Every new feature creates another question.

Every new data source creates another problem to solve.

Every improvement to the scanner creates another edge case hiding somewhere in a database wearing a tiny disguise.

And every time I think I've finally understood the technology stack, someone invents another technology stack.

Wonderful!

Exactly what I needed.

But that is also the point.

After redundancy knocked the wind out of my sails, I wanted something that gave me a reason to keep learning.

SavvySipper has done that in spades.

I've learned more about software development, databases, AI, deployment, automation, APIs, email infrastructure and product design in the last several months than I could ever have imagined when this started.

And I have learned most of it because something was broken.

Which is, apparently, a perfectly legitimate educational strategy.

From a Wine List in London Bridge to a Bigger Idea

What started with a slightly strange pricing anomaly on a restaurant wine list has grown into something much bigger than I originally imagined.

SavvySipper is becoming a platform for wine discovery, wine pricing intelligence and better decision-making.

For consumers, the goal is simple: help you drink better and spend smarter.

For restaurants, it is about understanding what customers see when they open the wine list and identifying opportunities to improve value without destroying profitability.

For retailers and wholesalers, it is about connecting the right wines with consumers and venues at exactly the point where interest and purchase intent are highest.

And for me?

It's about building something I believe should exist.

There Will Probably Be Another Disaster Tomorrow

There almost certainly will.

A database will misbehave.

A deployment will fail.

A button will move six pixels to the left for absolutely no reason.

An AI agent will confidently suggest a solution that makes things considerably worse.

I will stare at it.

I will mutter something inappropriate.

I will ask for help.

And then I will fix it.

That, in a slightly ridiculous nutshell, is what building SavvySipper has been like.

It has been part wine project, part technology education, part business experiment and part ongoing argument with a computer.

But most importantly, it has given me something I was looking for after a fairly brutal career curveball: a reason to build, learn, experiment and keep going.

So, Why SavvySipper?

Because wine shouldn't require a PhD in pricing psychology to understand whether you're getting decent value.

Because restaurant wine lists can be brilliant but unnecessarily confusing.

Because consumers deserve better information.

Because restaurants deserve useful data rather than another platform shouting at them.

Because retailers and wholesalers should be able to benefit when a diner discovers a wine they genuinely love.

And because, frankly, there is something deeply satisfying about taking an idea scribbled in your head and slowly turning it into something that actually works.

Sometimes.

Usually.

Eventually.

The Savvy Verdict

SavvySipper started with a wine list, a pricing anomaly and one fairly persistent question: why doesn't this make sense?

It became a much bigger question.

What if we could use technology, AI and better data to make wine easier to understand for everyone?

That is what I'm building.

It has involved wine, databases, OCR, AI, GitHub, Vercel, Supabase, Resend, Google Antigravity, an unreasonable amount of caffeine and more conversations with coding assistants than I would care to admit publicly.

It has also involved a fair amount of swearing.

But we're getting there.

If you are interested in where this slightly mad experiment goes next, keep an eye on The Pour.

There will be wine.

There will be data.

There will probably be another technical catastrophe.

And, hopefully, there will be considerably more bottles where the price actually makes sense.

Cheers.

Stephen. SavvySipper Founder