Q&A with Miguel Hernandez, Creator of the PaloCam

Predictable Designs Podcast · 21 Apr 2021 · 47m · Episode 29

What this episode covers

Miguel Hernandez, creator of the PaloCam, a pole-climbing camera for sports, discusses his journey with host John Teel, who analyzes Miguel's financials and technical design. Hernandez has sold 10 units at a loss, spending $680 per unit against a $275 sale price, due to high costs from 18 custom 3D-printed parts and 10 hours of manual assembly labor. Teel advises reducing part count via injection molding or urethane casting, outsourcing sub-assemblies to Asian suppliers like Seeed Studio, and using manufacturing partners to finance tooling and inventory through amortization and favorable payment terms.

Other shows on the same ground, closest first.

  • Thoroughly Considered · 31m

    17: Three Honda Civics

    Tom Gerhardt and Dan Provost of Studio Neat discuss the production of the Glyph tripod, detailing the $63,000 cost of injection molding tooling and the manufacturing process. They explain how they manage supply chain logistics for components like the wrist strap and wood handle while navigating the high capital requirements of hard tooling. The hosts also analyze the strategic impact of Apple product announcements on their business, noting how rumors of new iPhone camera features drive product development and timing for accessories.

  • Product Startup · 49m

    005: Miniature Construction Materials

    Mat Hofma and Erik Polumbo, founders of Mini Materials, explain how they launched a business selling miniature cinder blocks and pallets using a budget of under $500 for prototyping. They detail their process of 3D printing, creating silicone molds, and casting real concrete, while leveraging Instagram and Reddit for organic marketing. The duo discusses their decision to keep manufacturing in the U.S. to maintain margins and quality, and their strategy for scaling production by contracting local individuals rather than using large overseas factories.

  • Making It · 1h 01m

    Episode 293: My pants are here!

    Jimmy Diresta, Bob Clagett, and David Picciuto discuss recent projects, including DirestaWear pants, disposable shop knives, and 3D-printed helmets. Diresta details the high cost and low margins of manufacturing his DirestaWear pants in Manhattan, explaining why he prefers licensing his name over making clothes. Clagett demonstrates creating a disposable shop knife using an Elegoo resin printer and silicone molds, while also addressing technical issues with resin curing. Picciuto shares his experience selling out of books and designing impractical wooden boxes, emphasizing the value of 'impractical' projects for creative growth.

  • Being an Engineer · 40m

    S7E26 Paul Vizzio | From Prototype to Product: How Paul Vizzio Engineered RemieDog Into a Real Hardware Business

    Paul Vizzio, Director of Hardware Engineering at Proteus Motion and founder of the direct-to-consumer brand RemieDog, details the journey of engineering a specialized dog leash from a blocky 3D-printed prototype to a manufacturable product. He explains how he reduced costs by 80% by shifting from CNC machining to die casting and sourcing components from non-traditional hiking suppliers to ensure load ratings. Vizzio also discusses the harsh realities of unit economics, revealing that while the hardware is profitable, marketing and operational expenses have made the venture a financial loss, leading him to adopt a 'build in public' strategy to share his engineering and business learnings.

  • The Hardware Entrepreneur · 36m

    #032 - Why manufacturing in USA vs China is becoming interesting, with Scott N. Miller of Dragon Innovation, USA

    Scott N. Miller, co-founder and CEO of Dragon Innovation, discusses the shifting landscape of manufacturing between the US and China, highlighting how rising Chinese labor costs and the rise of crowdfunding are making US production viable for lower-volume hardware startups. He details Dragon Innovation's new software tool, Product Planner, designed to help hardware companies manage bills of materials and unit economics, and announces a partnership with Kickstarter and Avnet called Hardware Studio to support creators. Miller also shares a critical failure story from his time at iRobot involving the first 15,000 Roombas, where a design flaw caused optical encoders to fail due to rubber dust, illustrating the importance of testing and quality control.

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by Predictable Designs Podcast. The episode itself belongs to its publisher.