Bull @bull_gobull - Twitter Profile

Pinned Tweet

4 days ago

Serenity @aleabitoreddit has been digging into Korea’s AI supply chain. But some of the most interesting names may simply be too small to talk about publicly. I went down the rabbit hole and found 7: $90M–$1.9B market caps. HBM. Photonics. Testing. Materials. Some are already on his radar. Others fit the same supply-chain thesis. Here’s the map 🧵👇 First, the names Serenity has actually touched: 🔥 $138080 OE Solutions — ~$0.25B Photonics / EML / CW Laser 800G/1.6T, EML and CW lasers → direct exposure to AI data center optics and CPO. → Serenity disclosed a position. 🔥 $093370 Foosung — ~$1.05B WF₆ / Specialty Gas WF₆ is critical for tungsten deposition. The thesis: memory expansion + potential supply constraints = bottleneck exposure. → Direct Serenity research; no position disclosed. 🔵 $322310 AUROS — ~$0.09B HBM / Advanced Packaging Metrology Hybrid bonding requires increasingly precise overlay control as packaging complexity rises. → Serenity previously owned it, then reduced exposure as hybrid-bonding timelines slipped. Then there are the names further down the same supply-chain rabbit hole: 🟡 $104830 Wonik Materials — ~$0.30B Specialty Gases Samsung + SK Hynix supplier. Advanced DRAM/NAND → higher specialty-gas consumption. 🟡 $089030 Techwing — ~$1.29B HBM Testing HBM Cube Prober exposure. More stacked layers → greater testing complexity + higher yield-control value. 🟡 $357780 Soulbrain — ~$1.89B Semiconductor Chemicals High-purity etching and cleaning chemicals. More complex memory processes → greater material intensity per wafer. ⚪ $317330 Duksan Techopia — ~$0.19B Semiconductor Precursors HCDS, SiCl₄ and other deposition materials tied to DRAM/NAND capex. The real hunt isn’t for “the next SK Hynix.” It’s one layer deeper: AI → HBM → Advanced Packaging → Testing → Materials → Photonics Then look for the intersection of: Small Market Cap × AI Revenue Inflection × Supply Bottleneck The best picks-and-shovels businesses often sell something that represents only a tiny fraction of the final chip’s cost. But if that component becomes scarce, billions of dollars of downstream capacity can get held up. That’s what makes Korea’s AI semiconductor small-cap supply chain worth digging into. 🔥 Direct — explicitly researched / owned 🔵 Weakened — researched, but thesis weakened 🟡 Thematic — closely aligned with the research thesis ⚪ Candidate — supply-chain extension

$bull_gobull's tweet photo. Serenity @aleabitoreddit has been digging into Korea’s AI supply chain. But some of the most interesting names may simply be too small to talk about publicly. I went down the rabbit hole and found 7: $90M–$1.9B market caps. HBM. Photonics. Testing. Materials. Some are already on his radar. Others fit the same supply-chain thesis. Here’s the map 🧵👇 First, the names Serenity has actually touched: 🔥 $138080 OE Solutions — ~$0.25B Photonics / EML / CW Laser 800G/1.6T, EML and CW lasers → direct exposure to AI data center optics and CPO. → Serenity disclosed a position. 🔥 $093370 Foosung — ~$1.05B WF₆ / Specialty Gas WF₆ is critical for tungsten deposition. The thesis: memory expansion + potential supply constraints = bottleneck exposure. → Direct Serenity research; no position disclosed. 🔵 $322310 AUROS — ~$0.09B HBM / Advanced Packaging Metrology Hybrid bonding requires increasingly precise overlay control as packaging complexity rises. → Serenity previously owned it, then reduced exposure as hybrid-bonding timelines slipped. Then there are the names further down the same supply-chain rabbit hole: 🟡 $104830 Wonik Materials — ~$0.30B Specialty Gases Samsung + SK Hynix supplier. Advanced DRAM/NAND → higher specialty-gas consumption. 🟡 $089030 Techwing — ~$1.29B HBM Testing HBM Cube Prober exposure. More stacked layers → greater testing complexity + higher yield-control value. 🟡 $357780 Soulbrain — ~$1.89B Semiconductor Chemicals High-purity etching and cleaning chemicals. More complex memory processes → greater material intensity per wafer. ⚪ $317330 Duksan Techopia — ~$0.19B Semiconductor Precursors HCDS, SiCl₄ and other deposition materials tied to DRAM/NAND capex. The real hunt isn’t for “the next SK Hynix.” It’s one layer deeper: AI → HBM → Advanced Packaging → Testing → Materials → Photonics Then look for the intersection of: Small Market Cap × AI Revenue Inflection × Supply Bottleneck The best picks-and-shovels businesses often sell something that represents only a tiny fraction of the final chip’s cost. But if that component becomes scarce, billions of dollars of downstream capacity can get held up. That’s what makes Korea’s AI semiconductor small-cap supply chain worth digging into. 🔥 Direct — explicitly researched / owned 🔵 Weakened — researched, but thesis weakened 🟡 Thematic — closely aligned with the research thesis ⚪ Candidate — supply-chain extension$

Serenity

@aleabitoreddit

4 days ago

I’ve actually entered positions in many tiny Korean companies recently… That actively supply memory, test, machines, etc. to $MU, $SKHY, Samsung, and others like $SMTC + US companies. It’s pretty interesting what you find in the $30-$250m range for optical components, CPUs, memory, and others that are inflecting due to AI. I don’t share these names publicly since risks/volatility is enormous. But one or two going from $40M -> $400m makes the risk/reward worth it for me personally.

90

1K

47

364

420K

0

1

0

506

Bull

@bull_gobull

Last Seen Users on Sotwe

Trends for you

Most Popular Users