Selected Works
From the first satisfying sketch to factory floor — I design products built to be manufactured, not just rendered.
Selected Works
All concept direction, design decisions, and final deliverables are my own. AI tools are integrated throughout my process — see the AI-Assisted Workflow section for details.
About Me
7+ years as an industrial designer, solving problems across automotive, lighting, and consumer goods. My edge? I don't stop at the CAD file. I've walked factory floors in China, coordinated 50+ international suppliers, and learned what separates a beautiful concept from something that can actually be made.
Currently at ABDI Design in Toronto — leading bespoke automotive projects including Xolaris Cybertruck: 50+ CNC-machined components, delivered from concept to production in 8 months.
Tools: Rhino, SolidWorks, KeyShot, Illustrator, Photoshop
Approach
I work across the entire product lifecycle — from early concept sketches through prototyping to production-ready deliverables. With hands-on experience coordinating international manufacturing and a deep understanding of prototyping methods beyond 3D printing, I design products that are built to be made, not just rendered.
Across prototyping, I work from rough form studies in FDM up through production-representative parts in SLA, SLS, and metal 3D printing. For low-volume runs I specify vacuum casting, silicone molding, CNC machining, and sheet metal fabrication; for structural work, composite layup. I've specified finishes including PVD coatings, anodizing, powder coating, and specialty paints — choosing the method against what the part has to prove at each stage, not what's cheapest to order.
Direct relationships with manufacturers in China, developed through factory-floor visits and recurring production runs. I handle supplier selection, tolerance and material negotiations, sample approval, and the in-production problem-solving — wrong batch tolerances, substitute materials, finish defects — that sits between an approved design and a shipped unit.
DFM starts at the first sketch, not at the CAD review. Material choice, wall thickness and draft, tolerance stacks, assembly sequence, and per-unit cost are decisions that shape the form itself. I've designed parts for runs as small as 50 and as large as 10,000 — each scale forces different tradeoffs, and the design has to know which it's built for.
AI has compressed several stages of my design process without replacing the judgment that makes them work. Here's where each tool sits in my pipeline, what it does, and what it solves.
Tools: Claude (primary), ChatGPT. Application: Synthesizing user research notes, drafting design briefs, generating target persona language, and structuring technical documentation and BOMs. Problem solved: What used to be a full day of writing and organizing compresses to an hour of review and refinement, freeing time for the actual design decisions.
Tools: Midjourney, Krea, Nano Banana. Application: Running 40–60 directional images with controlled parameters (style weight, aspect ratio, stylization) to establish aesthetic direction before committing CAD time. Problem solved: Compresses the fuzzy front end from days of sketching to hours of guided exploration. Stakeholders weigh in on direction while it's still cheap to change.
Tools: Vizcom. Application: Converting hand and digital sketches into refined visuals. I use Vizcom's Drawing Influence control at 100% for form fidelity and lower values for prompt-led variation, exploring proportions while preserving design intent. Problem solved: Review cycles with non-designer stakeholders happen earlier — before CAD time is committed. Variations that used to take days of re-rendering now take minutes.
Tools: Vizcom, Nano Banana Pro. Application: Running 5–10 material and finish variations on the same form — anodized aluminum, powder coating, textile wraps, composite layups — to pressure-test CMF direction before specifying. Problem solved: CMF decisions that used to require physical samples or lengthy render queues now happen in a single review session.
Tools: Nano Banana Pro, Kling, Higgsfield. Application: Photorealistic scene rendering for stakeholder sign-off and marketing context. Short product reveal videos and orbiting camera sequences from single product images — applied when the pitch needs motion, not just stills. Problem solved: Final-stage client reviews happen days earlier in the production schedule. When the deck calls for video, I can turn a product render into a 5–10 second clip without hiring a videographer.
The tools compress the loop between intent and visible result. They do not replace the judgment about what to make, or the technical craft of making it.
Let's build something.
Toronto, Canada
