Ultra Clean Holdings

Phase A — Understand the business

Lens 1 · Company Overview

UCT is "a leading developer and supplier of critical subsystems, components, parts, and ultra-high purity cleaning and analytical services primarily for the semiconductor industry". It reports two segments:

• Products (87.6% of FY25 revenue, $1,799.3M): chemical/gas/fluid delivery modules, frame assemblies, precision robotics, process modules and high-level assemblies for wafer-fabrication equipment (WFE) and sub-fab support. This is fundamentally an outsourced integrated-manufacturing offering — UCT lets an OEM hand off design-to-delivery of complex subassemblies.
• Services (12.4%, $254.7M): ultra-high-purity parts cleaning, tool-part recoating, surface encapsulation, and micro-contamination analytics for device makers and WFE OEMs — higher margin (27.7% gross vs 14.0% in Products).

Business model. Standard arrangement = signed PO/contract, no right of return, payment 30–90 days; revenue recognized at shipment (products) or completion (services). Crucially, there is no take-or-pay and no material backlog: "we generally do not have a material backlog of unfilled orders and because of the short time frame within which we are often required to design, produce or deliver". UCT is a demand-taker with near-zero forward visibility — it flexes with its customers' order flow.

Customers / suppliers / competitors. ~95.7% of FY25 revenue came from the semiconductor industry. Two customers — Applied Materials and Lam Research — each >10% of revenue every year, together 58.7% of FY25 revenue (see Lens 4). Competitors span three fronts (Lens 3). FY25 total revenue $2,054.0M, −2.1% YoY.

Scale: 7,411 employees (463 temporary) as of 2025-12-26 — a labor- and working-capital-heavy manufacturer, not an asset-light IP licensor.

Lens 2 · Supply Chain

Position in the chain: UCT sits one tier below the WFE OEMs.

Component suppliers (valves, gas/fluid parts, robotics, weldments, chemicals/gases) → UCT (subsystem integration + parts cleaning/coating/analytics) → WFE OEMs: Lam Research, Applied Materials, ASML → Chipmakers: TSMC, Samsung, SK Hynix, Micron, Intel, and China domestic fabs → AI/HBM/logic end demand.

• Upstream inputs: purchased materials dominate Products cost (a $33.3M material-cost swing drove the FY25 Products cost decline). UCT also makes many of its own gas-delivery components (valves, connectors, thermal products, weldments) — a degree of vertical integration via acquisitions (Ham-Let gas components; HIS Innovations, acquired 2023-10-25 for 100% of shares, Hillsboro OR).
• Chokepoints / single-source: the filing flags exposure to sole/limited-source suppliers and to export-license timing; and a manufacturing footprint concentrated in low-cost Asia (Singapore, Malaysia, China), Czechia, and Israel. The Israel operations (gas components) carry geopolitical/continuity risk explicitly cited (Israel–Hamas conflict).
• Cleaning-services node = Cinos Korea/China (the JV, see Lens 9) — outsourced precision-parts cleaning/recycling physically located next to Korean device makers.
• Manufacturing is deliberately placed "close to customers and their end users" in low-cost regions — the diverse footprint is the cost-competitiveness argument.

Names present → lens satisfied. The single most important chokepoint is not a supplier — it is customer concentration on the demand side (Lens 3/4).

Lens 3 · Competitive Advantages (moats)

Named competitors:

• Gas delivery systems: Ichor Systems (ICHR), Fujikin
• Other critical subsystems: Foxsemicon, Jabil (JBL), VDL ETG, Celestica (CLS), Benchmark Electronics (BHE)
• Gas-delivery components: Swagelok, Parker Hannifin (PH), Fujikin, Watlow
• Services (cleaning/coating): KoMiCo, SHT, EnPro (NPO), Pentagon
• Analytical services: Balazs (Air Liquide), Eurofins, SCAS

How durable is the moat? Honestly — thin. UCT itself concedes competitors "have substantially greater financial, technical, manufacturing and marketing resources," expects "intensified price-based competition," and that "we may have to reduce the prices of our products". The real moats are modest and process-based:

1. Qualification / switching cost (weak-to-moderate): UCT has been "qualified as a supplier" with each customer per their qualification requirements; requalifying a new subsystem vendor into a production tool is slow. This is the strongest moat — but it is shared with Ichor and does not stop share shifts between qualified vendors (the Lam-share move in Lens 4 shows the door swings both ways).
2. Vertical integration + design-to-delivery cycle-time — a cost/convenience advantage, not a monopoly.
3. Services recipes (cleaning/coating process IP, historical relationships) — the higher-margin, stickier side.

Bargaining power: unfavorable. UCT needs Lam and Applied far more than they need any single subsystem vendor. The gravest structural risk in the filing is that the OEMs "could elect in the future to develop and manufacture these subsystems internally" (vertical in-sourcing). UCT is a margin-taker: 15.7% gross margin (Lens 5) is the tell — this is a supplier without pricing power.

Lens 4 · Segments (revenue / margin by product & geography)

All figures.

By segment (FY, $M):

Segment operating losses in FY25 are distorted by the $151.1M goodwill impairment (Q2-25: $77.6M to Fluid Solutions [Products], $73.5M to Services). Ex-impairment, both segments were modestly profitable. Services' 27.7% gross margin is nearly 2x Products' 14.0% — the mix-shift toward Services is the margin lever management is leaning on, but Services is only 12.4% of revenue and grew just +4.4%.

By geography (FY, $M): US $495.4M (24.1%) / International $1,558.6M (75.9%). International share has climbed every year (69.6%→73.0%→75.9%) as production shifted offshore. China domestic business is <5% of total revenue per management — important context for the export-control and securities-litigation risk (Lens 10).

Trend: FY24 was the peak (+20.9% up-cycle year); FY25 decelerated (−2.1%) even as industry WFE grew ~11% — i.e., UCT under-grew the cycle in 2025, a share/mix red flag. Q1-26 re-accelerated (+2.9% YoY, Lens 5).

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print — Q1 FY2026, quarter ended 2026-03-27)

All unless tagged.

• Beat vs. miss: Non-GAAP EPS $0.31 beat consensus by ~15%. Revenue in line-to-above. Stock +~8% on the 2026-04-28 print.
• The killer line is tax. UCT earned $11.4M operating income and $4.2M pre-tax — then booked a $19.2M tax provision (incl. a $14.8M discrete/deferred charge) driven by "losses in jurisdictions with full valuation allowances" and geographic earnings mix, turning a profitable quarter into a $(0.40) GAAP loss. This is structural, not one-off: the US has a full valuation allowance ($104.2M at FY25) so US losses generate no benefit, while profitable foreign subs pay cash tax. Cash taxes $6.1M in the quarter.
• Margins: GAAP gross margin slipped to 15.8%; management cites non-GAAP GM 16.5% (up from 16.1% prior quarter) and guides continued improvement. This remains a low-teens/mid-teens gross-margin business.
• Balance-sheet flags (Q1-26):
• Inventory built $91.0M in one quarter ("higher production levels") — a huge working-capital draw that swung operating cash flow to $(33.3)M (from +$28.2M Q1-25). With capex $9.6M, Q1-26 free cash flow ≈ $(42.9)M. Management frames the inventory as a demand ramp; a skeptic reads a $91M inventory build into a no-backlog business as a demand-timing risk (Lens 13).
• Major recapitalization (2026-03-03): issued $600.0M of 0.00% (zero-coupon) Convertible Senior Notes due 2031, repaid ~$462M of the secured Term Loan B, bought a capped call ($25.1M), and repurchased $40.0M of stock (0.7M shares). Weighted-avg borrowing rate cut from ~6.2% to ~1.4%. Conversion price ~$84.75 (42.5% premium over the $59.47 close on 2026-02-26); capped-call cap $104.07. Cash-interest burden roughly eliminated; dilution deferred until >$84.75 (mitigated to $104.07). Smart, opportunistic — and a tell that the stock was cheap in February.
• Guidance (Q2-26): revenue $565–605M (mid $585M, +~13% YoY / +~10% QoQ), GAAP EPS $0.20–0.36, non-GAAP EPS $0.44–0.60 (mid $0.52) — a sharp sequential ramp. Management "raised Q2 revenue outlook."

Net read: operationally inflecting up (revenue re-accelerating, guide strong, interest cost slashed) — but GAAP still loss-making because of tax, and cash generation went negative on the inventory build.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on disk (transcripts=0); grounded.

Management tone under new CEO James Xiao is markedly confident/bullish — a clear shift from the defensive, China-demand-softness messaging of early 2025. Framework rolled out as "UCT 3.0": (1) Ramp readiness ("execute with urgency, customer-first"), (2) MPX / Multi-Product eXpansion (regional centers of excellence US/EU/Asia to compress new-product-introduction cycles), (3) Digital transformation (AI-enabled operations). Recurring phrases: "multi-year cycle," "multiyear growth tailwind," "disciplined execution will define the winners," "well positioned to capture an outsized portion of opportunities."

Substance to weigh:

• WFE claim conflict (surface, don't resolve): management cited WFE "$140–145B in 2026, 18–20% YoY growth". SEMI's official Dec-2025 forecast is WFE +9.0% in 2026 to ~$135B (total equipment $139–145B; $156B in 2027). Management's 18–20% is materially more bullish than SEMI's WFE line — likely conflating total-equipment growth or citing a house bull case. Do not take the 18–20% at face value.
• "$4B revenue by 2030" (Vision 2030) with "modest incremental capital," 6–9 months to build capacity; "we do not expect infrastructure capacity to be our constraint". Management characterizes a "~$3B run-rate" today — but FY25 was $2.05B and the Q2-26 guide annualizes to ~$2.34B, so "$3B run-rate" is forward/aspirational framing, not a reported number (flag for Lens 13).
• Demand: "all major memory customers investing in greenfield factories"; leading-edge logic + HBM are "etch and removal intensive" — genuinely favorable for gas/fluid-delivery content per tool.

Sentiment trajectory: Feb-2025 (prior regime) = "demand softness in China, extended qualification timelines, inventory absorption" → stock −28% in a day. 2026 (new regime) = unbroken bullishness on an AI multi-year cycle. The tone has completely reversed; the business moved less than the tone.

Lens 7 · Comps

Peer set = semicap subsystem/component suppliers. Multiples `` per source/date; 5-yr avg ROE not sourced → "n/a" rather than fabricated.

Reads:

• Trailing/EBITDA multiples are near-useless right now — UCT (44x) and Ichor (127x) EV/EBITDA reflect trough, impairment-depressed EBITDA, not rich earnings. UCT TTM EBITDA is only ~$118.4M. The honest lens is EV/Sales and forward P/E.
• UCT at 2.52x EV/Sales and ~33x fwd P/E is expensive versus its own history (a low-margin subsystem CM has historically traded ~0.8–1.5x sales / high-single-to-mid-teens forward P/E). The re-rate is almost entirely P/S multiple expansion.
• Versus the higher-quality peers (MKSI 24.8x fwd P/E / 25x EBITDA, ENTG 24x EBITDA) — companies with proprietary tech and structurally higher margins — UCT and Ichor now trade at a premium forward P/E despite being the lowest-margin, most cyclical, least-moaty names in the group. That inversion is the valuation flag.
• Ichor (net-cash, +318% in 52 weeks) confirms this is a sector-wide small-cap semicap melt-up, not a UCT-specific fundamental story.

Lens 8 · Stock-Price Catalysts (moves >5%, last ~5 years)

Mostly; pattern-level.

• 2022 up-then-down: record FY22 revenue (~$2.4B) but a Q4-22 guide cut ($600–650M → $560–570M) on a weakening cycle + China COVID shutdown — the classic cyclical top.
• 2022–2023 downturn: revenue fell to $1.73B (FY23), net loss $(0.70)/sh; stock de-rated hard into the WFE trough.
• 2024 recovery: revenue +21% to $2.10B, back to $0.53 EPS — stock recovered.
• 2025-02-24 — the −28% day: disclosed China "demand softness / extended qualification timelines / inventory absorption"; stock fell $36.06 → $25.90 (−28%) in one session. Spawned the securities class action (Lens 10). CEO Scholhamer resigned weeks later.
• 2025 mid-year: sustained share-price decline triggered the $151.1M goodwill impairment.
• 2026 melt-up: ~+110% from 2026-03-31 to 2026-07-01; ~$59 (Feb) → ~$106 (Jul 9). Q1-26 beat (+8% on 2026-04-28), convert recap de-risked the balance sheet, and an AI/WFE + Vision-2030 narrative drove multiple expansion. Broker upgrades (Oppenheimer → $115, Outperform).

What the tape reveals: UCT is a high-beta WFE-cycle proxy that reacts to (1) WFE-cycle inflections, (2) its own guidance (esp. China/customer-demand signals), and (3) sector-wide narrative re-rating. Single-day ±8–28% moves on demand/guidance surprises are normal — the downside gaps (−28% Feb-25) are as violent as the upside. This is not a low-volatility compounder.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO — James (Jinsong) Xiao, since 2025-09-02 (age 55). Ex-Applied Materials Corporate VP/GM (Semiconductor Dielectric ALD + Metal Deposition/ALD BU); at AMAT June 2006–Sept 2025. Track record = ran multi-billion-dollar BUs at the customer. Note the optics: the new CEO comes from Applied Materials, a top-2 customer (23% of revenue) — deep domain credibility, but a governance eyebrow-raise on independence/related-party sensitivity.
• COO — Robert Wunar, appointed 2026-01-25 (effective 2026-03-23), also ex-Applied Materials (Managing Director, Business Unit Operations, to Jan-2026). The top two operating roles are now both ex-AMAT.
• CFO — Sheri Savage (long-tenured; UC Davis econ) — retained through the CEO transition and named as a defendant in the securities class action (Lens 10). A CFO carried through a leadership overhaul while under active investor litigation is a real governance overhang.
• Prior CEO — Jim Scholhamer resigned "effective immediately" in March 2025 for personal health reasons, weeks after the −28% China-demand day; founding CEO/Chairman Clarence Granger bridged as interim. FY25 G&A absorbed a "separation payment made to the prior CEO".
• Skin in the game — modest. Heavily institutional (95%); true insider/management ownership is low ($69M ≈ ~1.8% per Simply Wall St; a separate 46% figure appears to be a 13F/data artifact and includes 5%+ holders). Francisco Partners cited as a large (~13.8%) holder. This is a professionally-managed small-cap, not founder-led — no large aligned equity stake driving decisions.
• Capital allocation: mixed. Grew via debt-funded M&A (Ham-Let, HIS 2023) that then required a $151.1M goodwill write-off — value-destructive on the Fluid Solutions/Services deals. Buybacks are opportunistic and small ($40M in Q1-26 at ~$57, $29.4M in 2023, ~$0 in 2024). The 2026 convert recap was genuinely good capital allocation — cut cash interest ~6.2%→1.4%, extended maturity to 2031, added dry powder. No dividend.
• Archetype: turnaround/professional-operator regime installed to execute a growth-and-margin agenda ("UCT 3.0"), backed by customer-side pedigree. Bull: exactly the operator you want for a scale-up. Bear: unproven at UCT, thin ownership, and inheriting a litigation and margin problem.

Lens 10 · Forensic Red Flags

Grounded in the financials + for litigation.

Accounting / earnings-quality flags:

1. Cash conversion is weak and flattered. FY25 operating cash flow $65.6M was propped by two non-durable items: the $151.1M non-cash goodwill impairment add-back and $68.0M of receivables sold via non-recourse factoring ($56.4M + $11.6M Fluid Solutions, the latter cancelled Dec-2025). Factoring pulls AR off the balance sheet and into operating cash flow — the $32.3M AR "release" that helped FY25 OCF is substantially manufactured, not earned. FY25 FCF ≈ $15.3M on $2.05B revenue = 0.7% FCF margin. Q1-26 FCF ≈ $(42.9)M.
2. Inventory outrunning revenue. Inventory rose to $390.9M (22.6% of total assets) at FY25 while revenue fell 2.1%; then +$91.0M more in Q1-26. In a no-backlog business, inventory building ahead of demand is the single most watchable line — bullish if the ramp is real, a write-down risk if it isn't.
3. Tax turns profits into losses. Full US valuation allowance ($104.2M) + foreign cash taxes + incoming Pillar Two (Singapore/Malaysia low-tax incentives erode from FY26) mean GAAP net income will lag operating income for the foreseeable future. Non-GAAP EPS flatters the picture; GAAP was a loss in FY25 and Q1-26.
4. Goodwill/intangibles still sizeable post-impairment: goodwill $114.2M (mostly Core Products) + intangibles $156.8M = $271M, ~16% of assets — further impairment risk if the Core Products fair value cracks (a critical audit matter flagged by PwC).
5. SBC modest ($19.2M FY25, ~0.9% of revenue) — not a major non-GAAP distortion here.
6. Leverage / rating: post-recap gross financial debt ~$601.9M vs cash $323.5M → net financial debt ~$278M; net debt / ~$118M adj. EBITDA ≈ 2.3x, ~3.9x including operating leases. Term-loan pricing was tied to a Ba3/BB− (below-investment-grade) rating grid. Altman Z ≈ 2.51 (grey zone). Manageable, not pristine.
7. Auditor change: PwC since 2024 (Baker Tilly 2015–2024) — a mid-2024 auditor rotation; clean opinions, effective ICFR, no material weakness disclosed. Worth noting given the concurrent litigation window.

Regulatory findings (required sub-section):

• SEC (EDGAR EFTS — LR + AAER): No Litigation Releases or AAERs naming Ultra Clean Holdings, 2021-07-10 → 2026-07-10.
• Private securities litigation — MATERIAL, ACTIVE: a securities class action names UCT, former CEO James Scholhamer, and sitting CFO Sheri Savage, alleging materially misleading statements about Chinese-market demand to inflate the stock; class period 2024-05-06 to 2025-02-24; the 2025-02-24 "demand softness" disclosure drove the −28% drop. Case early-stage; management calls it meritless and can't estimate a loss. This is the key governance/legal overhang — a live case turning on the very China-demand narrative that has now flipped bullish.
• 10-K Item 3 (Legal Proceedings): boilerplate "various legal proceedings and claims… in the ordinary course… not… material" — does not specifically carve out the class action as material.
• Non-SEC enforcement (FTC/DOJ/etc.): no material hits found.
• Verdict: No SEC enforcement; one active, material securities class action on China-demand disclosures; otherwise clean audit/ICFR.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (EPS, next 3 fiscal years)

Bottom-up from Q1-26 actual + Q2-26 guide + WFE cycle. Outputs ``; not logged to forecast.ts (watchlist/unattended + wave boundary).

Anchors: FY25 revenue $2,054M; Q1-26 $533.7M actual, Q2-26 guide mid $585M; non-GAAP EPS Q1 $0.31, Q2 mid $0.52; WFE +9% (SEMI) to +18–20% (mgmt) in 2026.

Arithmetic note (FY26 base): 4 quarters ≈ $0.31 + $0.52 + $0.55 + $0.62 = $2.00. GAAP EPS will run well below non-GAAP because of the tax structure (Lens 5).

Valuation cross-check: at ~$106, that is ~53x FY26 / ~35x FY27 non-GAAP EPS. The stockanalysis "fwd P/E 33.5" implies the street is on ~$3.17 (≈FY27). The multiple only works if FY27–28 land AND the multiple holds — i.e., the market is paying a quality/secular multiple for a low-margin cyclical.

Base call (for later scoring, not logged): "UCTT FY2026 non-GAAP diluted EPS ≥ $1.90" — p ≈ 0.65 (Q1+Q2 already ~$0.83 mid; needs a normal H2).

Lens 12 · Bull vs Bear

Bull case. UCT is a direct-content beneficiary of the most gas/fluid-delivery-intensive part of the AI capex cycle — leading-edge logic + HBM/DRAM are "etch and removal intensive," and every incremental deposition/etch chamber Lam and Applied ship carries UCT subsystems. WFE is in a record multi-year up-cycle ($135B+ WFE, total equipment to $156B by 2027 ). The 2026 convert recap slashed cash interest (~6.2%→1.4%), de-risked the balance sheet, and extended runway to 2031. New ex-AMAT leadership brings customer-side credibility and a credible margin/scale agenda ("UCT 3.0," Vision 2030 = $4B revenue). Services mix + MPX could lift gross margin off the 15–16% floor. Earnings surprise vector: if H2-26 revenue runs $600M+/qtr and GM breaks 17%, FY27 non-GAAP EPS ~$3.30 makes today ~32x — defensible for a secular grower.

Bear case (permanent-impairment risks).

1. Customer concentration is existential. Lam (37.0%) + Applied (21.7%) = 58.7% of revenue, and Lam's share is rising. Any in-sourcing decision, share-shift to Ichor, or a single-customer capex air-pocket is a >10-point revenue event. The OEMs explicitly could build these subsystems internally.
2. No pricing power, structurally low margin. 15.7% gross / ~2% underlying operating margin / 0.7% FCF margin. This is a contract manufacturer; multiple compression to a CM multiple (0.8–1.5x sales) from today's 2.5x would roughly halve the stock regardless of earnings.
3. Tax + Pillar Two keep GAAP earnings suppressed and cash taxes high; the US valuation allowance means domestic softness generates no tax shield.
4. Expectations baked in: ~50x FY26 / ~35x FY27 non-GAAP for the lowest-quality name in the peer group. The +110% move was P/S expansion, not earnings.

Pre-mortem (18 months out, thesis broke): WFE growth normalized to SEMI's +7–9% (not management's 18–20%); a memory or China order pause hit Q3/Q4-26; the $91M inventory build became a write-down; and the multiple de-rated from ~35x to ~18x forward as the small-cap semicap melt-up unwound. Stock back to the $50–60s. The China securities class action survived a motion to dismiss, adding a headline overhang.

Contrarian view (what the market is refusing to see): the market is treating UCT as a secular AI compounder and pricing it like one, while the filing describes a price-taking, backlog-less, 15%-gross-margin subsystem CM that just under-grew its own cycle in 2025 and is being sued for overstating China demand. The tone reversed; the business model did not.

Lens 13 · Devil's Advocate (short-seller)

• What structurally breaks the model: OEM vertical in-sourcing, or Lam/AMAT dual-sourcing more aggressively to Ichor/Foxsemicon to de-risk their own supply — UCT has no contractual protection (no take-or-pay, no backlog).
• Revenue concentration: 58.7% in two customers; 17.1% of gross AR is Lam alone. Lose 5 points of Lam wallet-share and the growth story inverts overnight.
• Weaker moat than bulls think: qualification is the only real switching cost and it is shared with Ichor; UCT itself under-grew WFE in 2025 — proof the moat doesn't hold share in a shifting-demand environment.
• Most dangerous competitor: Ichor — the pure gas-delivery comp, net-cash, +318% in 52 weeks, taking the same AI narrative. In a price war for Lam/AMAT subsystem volume, the lower-leverage balance sheet wins.
• Worst capital-allocation history: debt-funded M&A (HIS/Fluid Solutions) written down $151.1M within ~18 months.
• Assumptions that MUST hold for $106: WFE +18–20% (management's number, not SEMI's); GM climbing to 17%+; no customer air-pocket; multiple staying ~35x forward; the class action fading. Break any one and downside is large — recall −28% in a single day on exactly this kind of demand surprise (Feb-2025).
• Growth disappoints 20–30%: FY26 revenue to ~$1.9B, non-GAAP EPS to ~$1.10; on a de-rated 18–20x that is a ~$20–25 stock — 60–75% downside. The asymmetry at $106 is unfavorable.
• Single permanent-impairment scenario: an OEM formally in-sources gas/fluid delivery (Applied has done deposition subsystems internally before) → structural TAM loss, not a cyclical dip. Plausibility: low-to-moderate, but non-trivial and un-hedgeable.

Lens 14 · Management Questions (ordered by information value)

1. Lam is now 37% of revenue and rising — what is your actual dual-source position with Lam by program, and what contractual protection exists against share-shift or in-sourcing?
2. The $91M Q1-26 inventory build — is it against firm customer demand signals/POs, and what is the write-down exposure if H2 orders slip?
3. Vision 2030 / "$4B revenue": what WFE assumption underlies it, and how do you reconcile your "18–20% WFE 2026" with SEMI's +9%?
4. Post-recap, what is the GAAP tax path — when (if ever) does the US valuation allowance reverse, and what is Pillar Two's FY26–28 drag on Singapore/Malaysia?
5. Gross-margin bridge to 17–18%: how much is mix (Services), pricing, or fixed-cost absorption, and is any of it durable if volume dips?
6. With both CEO and COO from Applied Materials (a 22% customer), how is the board managing related-party independence and pricing arm's-length-ness?
7. Status and reserves for the China-demand securities class action — and what disclosure controls changed since Feb-2025?
8. What share of the AI/HBM/leading-edge WFE uptick is UCT content per tool vs. cycle beta — where are you winning content, not just riding volume?
9. Capital allocation now that cash interest is ~1.4%: buybacks (you bought $40M at ~$57), M&A, or de-lever the convert before 2031?
10. Why did UCT under-grow WFE in 2025 (−2.1% vs +11% industry), and what specifically fixed it in Q1-26?
11. Services is your margin engine at 27.7% GM but only 12.4% of revenue and grew 4.4% — what is the realistic ceiling on Services mix?
12. Cinos Korea/China JV (the NCI): buy out minorities, or keep the structure — and what is its standalone growth/margin?
13. Factoring: $68M of receivables sold in FY25 — is this a permanent working-capital tool, and how should we read "underlying" operating cash flow without it?
14. Which single customer program, if cancelled, would most impair FY27 — and how concentrated is that risk?
15. If WFE flattens in 2027 (SEMI shows deceleration to +7.3%), what is the decremental margin, and how fast can the cost base flex given no backlog?

---