How I AI: Nicole Ruiz’s System for Buying High-Quality Goods and Automating Returns with Claude

Welcome back to How I AI! I’m Claire Vo, and I’m always on the hunt for practical ways to use AI to make life better. Today, I was so excited to sit down with Nicole Ruiz, a writer, parent, and the creator of a truly brilliant AI shopping system. If you’ve ever felt overwhelmed by the endless scroll of low-quality, drop-shipped junk online, this episode is for you. Nicole was tired of the panic-buying cycle that left her with plastic, poorly made items for her kids.

Like many of us, Nicole wants to fill her home with thoughtfully made items from natural materials—things that can be mended, cared for, and passed down. The problem is, the modern internet is actively working against that. It’s a minefield of paid ads, fake reviews, and knockoff brands. Instead of giving up and defaulting to Amazon, she decided to fight back with AI.

In our conversation, Nicole walks us through two incredible workflows. First, she shows us the household management project she built in Claude to act as her personal shopping assistant, one that vets brands for quality and heritage. Second, she reveals how she uses Claude Cowork to automate the entire annoying process of returning a faulty product, from finding the receipt to drafting the perfect refund request. It’s a complete system for managing the lifecycle of your purchases, reducing waste, and saving a ton of mental energy.

These are workflows you can start using today to shop smarter and reclaim your time from tedious digital admin. Let's get into it.

Workflow 1: Building an AI-Powered 'Buy It For Life' Shopping Assistant

The biggest challenge for any conscious consumer is noise. How do you find the small artisan or the century-old manufacturer when they’re being drowned out by flashy, direct-to-consumer brands with massive ad budgets? Nicole’s solution was to build a personal shopping filter inside a Claude project, creating a system that understands her values and does the research for her.

Step 1: Establish Your Purchasing Philosophy

Before you can delegate shopping, you have to know what you stand for. Nicole started by turning her own “invisible checklist” into a visible one. She created a dedicated project in Claude and fed it a very specific set of instructions that act as its core principles. This is the foundation of the entire system.

Her criteria include:

• Longevity and Repairability: Brands that have been in business for decades and whose products are made to last and be repaired.
• Trustworthy History: Brands sought out for their products for a long time, with a history of extraordinary craftsmanship.
• Strong Return Policies: Companies that stand behind their products and make returns easy.
• Avoiding the Hype: A specific instruction to avoid “trendy, direct-to-consumer brands that are paying a lot for advertising and probably underinvesting in quality.”

She even tells Claude to be skeptical of AI-generated reviews and to try and identify drop-shipping brands. By codifying these values, she ensures every recommendation is pre-filtered through her personal quality control.

Step 2: Curate Your Trusted Vendor List

Next, Nicole gave the AI a head start by providing her own list of trusted vendors. This came directly from a running list she kept in Apple Notes. Instead of starting from the entire internet, the AI now has a preferred list of places to look first.

Her list includes vendors like Boston General Store and Manufactum, which are themselves curators that vet high-quality items. This creates layers of trust; the AI is searching through stores that are already doing their own vetting.

Step 3: Define the Output Format

To make the results easy to compare, Nicole gave Claude a strict formatting rule for its output. For each recommended item, she wants to see:

1. Product Name
2. Photo
3. Price
4. Materials (surfaced prominently to avoid plastic)
5. Care & Maintenance Notes (e.g., "hand-wash only")
6. Purchase Link
7. A brief note on why the brand has a trustworthy history

This last point is my favorite part. It’s not just what to buy, but why it's a good choice, grounded in history and quality signals.

Step 4: Putting It to the Test

With the system in place, shopping becomes incredibly simple. Nicole showed me a few examples. First, she needed a can opener.

Her prompt was simply:

Help me find a can opener

Claude, using the project instructions, immediately searched her trusted vendors. It returned a few options, including the NuGen Super Kim can opener, available from both Boston General Store and Manufactum. It noted the brand was established nearly 100 years ago and is known for its kitchen tools. It also surfaced the price and materials right at the top. The best part? It found a high-quality item for the same price as a lower-quality one from a big-box store.

She also demonstrated how to use it for vetting new brands. She saw an ad for a brand and asked Claude to analyze it. The AI came back with a strong warning: "Do not add." It discovered the company had recently taken a private equity investment, and since then, reviews have been terrible. It flagged controversy over the CEO, poor internal Glassdoor reviews, and heavy spending on influencer marketing—all red flags for Nicole’s purchasing philosophy.

Workflow 2: Automating Product Returns with Claude Cowork

Even with the best vetting, some products fail. Nicole showed me a pair of J.Crew pants for her son that wore through in the seat after just a few months. For most of us, this is where a mountain of friction begins: find the receipt, look up the order number, find the customer service email, write a polite-but-firm message... it's a pain.

This is where Nicole’s second workflow kicks in, using Claude Cowork to handle the entire return process.

Step 1: Capture the Evidence and Start the Prompt

While tidying up, Nicole found the pants. Instead of adding it to a mental to-do list, she took action immediately. She took a photo of the worn-out pants and, using voice-to-text, kicked off a prompt.

Step 2: Dictate the Mission

While multitasking (she mentioned she might even be nursing a baby!), she gave Claude Cowork its instructions. This is a perfect example of using voice to get things done when your hands are full.

Here’s the prompt she dictated:

I have this pair of pants of Rafa's from the last six months they've already worn through in the butt. I'm looking to return them to J. Crew and specifically get a refund. Can you help me draft an email and specifically start out by finding the receipt for the pants in my email, either from PayPal or J. Crew with the item number and any other details you might need? Please include that context and draft up the request for refund to J. Crew customer service.

Step 3: The AI Assistant Gets to Work

This is where Cowork shines. Because it can connect to her tools, it immediately went to work on the tasks she assigned:

1. Searched her Gmail for receipts from J.Crew or PayPal related to the pants.
2. Extracted the key details: item number, order number, size, and purchase date.
3. Drafted a customer service email incorporating all that information.

Step 4: Review and Send the Perfect Email

Within minutes, Nicole had a perfectly crafted email ready to go. It included all the relevant order details in the subject line to prevent back-and-forth with the agent. It also included compelling language like, "The degree of deterioration is far beyond what I would expect from any garment at J.Crew's price point." It even found other reviews online mentioning the same quality issue, suggesting a manufacturing defect.

All she had to do was review it and click send. A 15-minute administrative task that would have been put off for days was completed in under two minutes, all while walking around her house.

Reducing Friction, Reclaiming Time

What I love about Nicole’s system is how it forms a complete loop. The shopping assistant helps bring high-quality items into her home, and the returns assistant efficiently deals with anything that doesn't meet the standard. Together, they radically reduce the mental overhead and administrative work of running a household.

This is a powerful demonstration of how AI can support a more thoughtful and sustainable way of living. It helps level the playing field for smaller artisans and heritage brands that, as Nicole noted, often have the worst websites. As our guest Jason Levin once said, “no UX is the best UX.” AI allows you to bypass those clunky interfaces and get right to the quality products.

By automating the busywork, Nicole gets to spend more time on the human parts of parenting. Her story is a fantastic inspiration to think about the invisible checklists and tedious tasks in our own lives and ask: could an AI do that for me?

Thanks to our sponsors

• Orkes—The enterprise platform for reliable applications and agentic workflows
• Metaview—The Agentic Recruiting Platform for winning teams

Episode Links

Where to find Nicole Ruiz:

• X: https://x.com/nwilliams030
• Substack (The Third Oikos): https://www.thirdoikos.com/

Where to find Claire Vo:

• ChatPRD: https://www.chatprd.ai/
• Website: https://clairevo.com/
• LinkedIn: https://www.linkedin.com/in/clairevo/
• X: https://x.com/clairevo

Tools referenced:

• Claude: https://claude.ai/
• Claude Cowork: https://www.anthropic.com/product/claude-cowork

Other references:

• Boston General Store: https://bostongeneralstore.com/
• LL Bean: https://www.llbean.com/
• Manufactum: https://www.manufactum.com/
• 5 OpenClaw agents run my home, finances, and code | Jesse Genet: https://www.lennysnewsletter.com/p/5-openclaw-agents-run-my-home-finances
• From a $6.90 newsletter to $3M API: How a non-coder built Memelord | Jason Levin: https://www.lennysnewsletter.com/p/from-a-690-newsletter-to-3m-api-how

Production and marketing by Penname. For inquiries about sponsoring the podcast, email jordan@penname.co.