Interview: Zoom Dustin Ross Skiffra — 2026-05-02
Key Themes
Operational data latency as the core problem. Skiffra’s founding insight is that enterprise systems like SAP generate the right data but deliver it to the wrong people, at the wrong time, and in the wrong interface. The gap between data capture and decision-making at the operator level is where value is lost — and where Skiffra intervenes.
Proof-of-concept results are striking. The Indonesian mining operation generated $100M in permanent annual EBITDA lift for ~$7M in fees. The truck-loading case alone ($30M EBITDA improvement via volumetric cameras + real-time weight sensing + a simple red/green interface) is a clean, compelling story about closing the last mile between data and the human decision-maker.
‘Hardware is hard’ as a moat. Andy explicitly cited this as a competitive differentiator — the physical interface layer, on-site compute requirements, and the need for operator-level UX create barriers that pure-software competitors can’t easily replicate. Skiffra is betting that the difficulty of the physical layer is a feature, not a bug.
Heavy industry as an underserved AI frontier. The team’s thesis is that software ate the easy problems first (all-digital workflows), and that atoms-meets-data industries — mining, oil & gas, manufacturing — are the next large surface area. Safety constraints have historically slowed adoption, which also means less competitive saturation.
Early-stage but credible pedigree. Founded early 2024, ~2 founders + 4 open engineering roles + ~10 international team members. Andy from Wall Street/natural resources, George from Toyota digital transformation. The team is lean but the domain expertise is unusually deep for a seed-stage company.
Notable Quotes
- “SAP is great at reporting, bad at predicting.”
- “We never think of the guy operating the loader as a decision maker — but he is.”
- “Hardware is hard — that extends to any physical interface. That’s a real frontier now.”
- “$7M in fees, $100M in permanent annual EBITDA lift.”
- *“You’re always going to need to get the rock through the processing and through the smelter. AI is not going to change that fact.”
Surprises
- The ROI ratio (14:1 on fees) is unusually high and credibly sourced — not a projection but a realized outcome from a single operation.
- The meeting was lower-context than expected: Andy joined solo (Alex and George unavailable), and neither party had been well-briefed on the other. The conversation was exploratory rather than pitchy.
- Skiffra is tracking film industry as a potential adjacency — surprising given the mining/oil & gas focus, though the framing (operational complexity + physical coordination) is coherent.
- The company appears to be pre-product in the sense of a repeatable software offering — the proof of concept was bespoke. The ‘stem cell team’ framing suggests they’re now moving toward productization.
Open Questions
- How does Skiffra’s model intersect with semiconductor supply chain specifically? The overlap with Project TBD is latent but not yet articulated.
- What is the actual product motion — is this land-and-expand consulting, SaaS, or outcome-based pricing?
- What does the Eleanor mine (Canada) engagement look like — is it a replication of the Indonesia playbook or a new capability?
- How does Skiffra think about data rights and IP from customer operations? (Federated model noted, but details thin.)
- Is there a world where Skiffra’s operational data layer is relevant to critical minerals supply chains that Project TBD tracks?