In one line: the nuclear fuel trade has worked and still has room, but the next leg is demand-led: Western reactors actually getting built. So own the tech-agnostic suppliers of long-lead parts (Doosan, BWXT) and Westinghouse via Cameco, and on the fuel side go to Canada, where the Athabasca juniors are cheaper than Cameco into a consolidation wave.
From policy to FID. Six years of rhetoric, policy and financing; the next five are about getting projects to final investment decision. The AP1000 pipeline is now 91 reactors, and Westinghouse (filed to IPO; Cameco 49%) is the best proxy. (2026-SEP-16)
No binary tech risk. Ocean Wall's 144-company, 25-point screen (deliverability + localization) lands on suppliers that are paid whichever design wins. Doosan is the only allied forger of large pressure vessels, and BWXT wins whichever SMR wins. (2026-SEP-16)
Golden screws. Long-lead equipment is the binding constraint. Siemens Energy and GE Vernova at 70–80x earnings on five-year sold-out order books show the premium once a bottleneck is discovered. (insights)
Canada, Canada, Canada. The ~50m lb/yr uranium deficit falls mostly on the West. He owns Cameco but calls it "not cheap," and sees better value in Athabasca juniors ahead of M&A. (2026-SEP-16)
The West's build gap. China builds for under $6bn and EDF is seconding engineers there. The drivers he names are regulation, NIMBYism, subsidies, labor and an "engineering state" (Breakneck); licensing reform and government first-loss capital for SMRs are the fixes. (2026-SEP-16)
The product
Grounded in his own description on 2026-SEP-16 (00:09–04:51).
What it is:Ocean Wall is a London firm founded ~7 years ago that publishes research "on all aspects of the nuclear fuel cycle." Examples are the report "From Blueprint to Fleet: Mapping the Nuclear Trade" and a stand-alone BWXT report that Finegold offers to share. With partners 5T in Zurich it also runs a dedicated 5T Nuclear Fund that carries "quite a large supply chain exposure." It is also putting together development capital for a UK SMR project.
Offering
What it is
How he runs it
Seen in the index
Research reports
Thematic and single-company nuclear research
Bill-of-materials mapping and a 144-company, 25-point supplier score (deliverability + localization), with lead-time tables and an RPV-forging matrix
Raising the capital infrastructure funds don't usually provide; argues government should fund the first $300–500m
—
How it serves retail investors: the supplier screen turns a hard-to-follow reactor build-out into a ranked list of listed names, including non-US ones like Doosan that retail rarely covers.
The tech-agnostic filter suits retail: it avoids having to pick the winning reactor design.
Transcripts
One dated page per appearance — each has its stock table, talking points, and the saved transcript. Newest first.