Some people say I look like Napoleon

But I don’t see it.

Old World Vulture opening for MogwaiPhoenix Concert Theatre, Toronto · April 26, 2011

Mike Costanzo in a black hat and plaid shirt, looking straight out with a slight smile.
But I don’t see it.
Old World Vulture opening for MogwaiPhoenix Concert Theatre, Toronto · April 26, 2011Watch on YouTube

Building an audience

The music was half the work. I built websites, designed artwork, booked shows and sold tickets. Making something great was only the beginning; finding its audience taught me how products reach people.

Early to the tools

I built a design team and challenged its tools. At Rangle I built a world class team of 15+ designers and pushed for rapid prototyping over static mock-ups. We adopted Figma during its invite-only preview; by launch, it was our design tool of choice (Lindsie Canton it was like yesterday).

Remote before it was normal

Remote teams. Close to users. At My Pain Sensei I built a remote product and engineering team before the pandemic. At Onist, I brought engineers into usability testing so they could see people using what they built—and build the right thing.

Strategist, then SVP

I sold the strategy and led the team through delivery. Back at Rangle, I moved from principal strategist to SVP, leading 15 people across product and design. In summer 2024, I prototyped Nike Kit, a GenAI shopping assistant, in under 30 days.

Blurring the lines

I’m closing the gap between design and delivery. At Nimble Gravity I worked as a product and design team of one. With Cursor and Claude Code, I moved from research and prototypes to shipping front-end code to production.

  1. 01 / 03 · 1 experiments

    Experiments

  2. CounterpartA contractor texts an order. AI handles the clear-cut lines and brings the tricky ones to a human. See how the review threshold works.

    Counterpart

    Section
    Experiments
    Type
    Live demo
    Year
    2026
    Open

    A contractor texts an order. AI handles the clear-cut lines and brings the tricky ones to a human. See how the review threshold works.

  3. 02 / 03 · 4 pieces

    Ideas

  4. The Big BottomThe next AI opportunity may be the customers everyone overlooked. Explore how agentic operations can make previously uneconomic markets viable.

    The Big Bottom

    Section
    Ideas
    Type
    Essay
    Year
    2026
    Read it

    The next AI opportunity may be the customers everyone overlooked. Explore how agentic operations can make previously uneconomic markets viable.

  5. Avoiding the Design System Death SpiralA component library can grow while its value shrinks. Lessons from years of building design systems on spotting the spiral and keeping a system useful.

    Avoiding the Design System Death Spiral

    Section
    Ideas
    Type
    Talk
    Year
    2024
    Read it

    A component library can grow while its value shrinks. Lessons from years of building design systems on spotting the spiral and keeping a system useful.

  6. Lean DiscoveryWhat if the next feature solves the wrong problem? A practical approach to mapping customer pain and testing solutions before committing to a build.

    Lean Discovery

    Section
    Ideas
    Type
    Article
    Year
    2023
    Read it

    What if the next feature solves the wrong problem? A practical approach to mapping customer pain and testing solutions before committing to a build.

  7. How Steve Jobs REALLY helped Apple get back to basicsOne Steve Jobs quote became an excuse to ignore customers. Revisit what he actually meant, and what Apple’s return to basics can teach product teams.

    How Steve Jobs REALLY helped Apple get back to basics

    Section
    Ideas
    Type
    Article
    Year
    2022
    Read it

    One Steve Jobs quote became an excuse to ignore customers. Revisit what he actually meant, and what Apple’s return to basics can teach product teams.

  8. 03 / 03 · 3 case studies

    Case studies

  9. Nike KitPrototyped in under 30 days, this personalized AI shopping assistant won executive buy-in and expanded Rangle’s work with Nike.

    Nike Kit

    Section
    Case studies
    Type
    Case study
    Year
    2024
    Open

    Pushing the boundaries of personalized shopping with a Gen-AI-enabled POC for one of the world's leading brands.

  10. MovebuddyFrom a stalled product to 5,000+ booked moves. How customer discovery, instant quotes, and a trust-first strategy brought a moving marketplace to life.

    Movebuddy

    Section
    Case studies
    Type
    Case study
    Year
    2023
    Open

    How to find product-market fit quickly, and take the stress out of moving.

  11. My Pain SenseiA patient-focused product pivot preceded a $30M acquisition within 18 months. See how research reshaped a tracking tool into a chronic-pain management program.

    My Pain Sensei

    Section
    Case studies
    Type
    Case study
    Year
    2018–2020
    Open

    Turning a failing product into a successful exit in less than 2 years.

Highlights

Smaller pieces, over the years

Smaller projects, experiments, and details from over the years. Select an image or video to explore the project.

The Faint Signal · Ask about Mike

Ask about Mike's work, how he thinks about product, or what he's looking for next.