Ichor Holdings

Phase A — Understand the business

Lens 1 · Company Overview

Ichor designs, engineers, and manufactures critical fluid-delivery subsystems — gas and chemical delivery systems — that go inside the wafer-fab equipment (WFE) tools its OEM customers sell to chipmakers. A gas-delivery subsystem is the precision plumbing of an etch or deposition tool: mass-flow controllers, regulators, pressure transducers, valves, and an integrated control system that delivers exact quantities of process gas. Chemical-delivery subsystems blend and dispense the reactive liquids used in CMP, electroplating, and wet-clean steps. Ichor also sells weldments, valves, its patented Advanced Flow Controller (AFC), and precision-machined components — increasingly its own proprietary parts rather than bought-in components.

Business model: Ichor is an outsourced design-and-build partner to WFE OEMs. It engages early in the OEM's tool-development cycle, co-designs the subsystem, and aims to be the sole-source supplier through the full life of that tool. No long-term volume contracts — it operates on a purchase-order basis. That is the key contract-structure fact: revenue is non-contracted and cyclical, but stickiness comes from qualification lock-in, not paper.

Customers (extreme concentration): Top customers in 2025 were Lam Research, Applied Materials, and ASML. Two customers over 10% — Lam Research and Applied Materials — were a combined 76% of total sales in 2025. Over 90% of revenue is semiconductor-capital-equipment-derived. Aerospace/defense is <10% of sales and pitched as the growth-diversification leg.

Scale: FY25 revenue $947.7M (FY24 $849.0M, FY23 $811.1M); ~1,891 full-time + 557 contingent employees; facilities in the US, Singapore, Malaysia, and Mexico after exiting Scotland and Korea in 2025.

Lens 2 · Supply Chain

Upstream → Ichor → end customer:

• Upstream inputs: mass-flow controllers, regulators, pressure transducers, filters, substrates, valves, fittings — sourced from "a large number of sources, including single-source and sole-source suppliers". Ichor uses supplier-consigned and just-in-time stocking. The historically dominant bought-in component is the MFC (the merchant MFC market is led by MKS Instruments, Horiba, and Brooks/Bruker). Ichor's strategic thrust is to replace bought-in MFCs with its own AFC and to in-source valves, fittings, and substrates — the proprietary-content shift below.
• Ichor (the integrator): clean-room assembly/test in Singapore, Oregon, Texas (Class 100 / Class 10,000); weldments in Malaysia, Oregon, Texas, California; chemical-delivery components in Oregon and Malaysia; precision machining + plating in California, Minnesota, and Mexico (the new substrate/valve insourcing hub); ITAR-compliant + AS9100 facilities in Minnesota and California for A&D.
• Downstream (the chokepoint): the WFE OEMs — Lam, Applied Materials, ASML — who in turn sell to TSMC, Samsung, Intel, Micron, SK Hynix. Five OEMs are >70% of all WFE revenue; Ichor sells into three of them. The single-source dependency runs the wrong way for Ichor: its customers are concentrated and powerful; its own supply is multi-sourced and managed JIT. The chokepoint that matters is customer concentration (Lens 3 / Lens 13), not input scarcity.

Lens 3 · Competitive Advantages (moats)

• Primary moat — qualification lock-in / sole-source position. Once Ichor co-designs a subsystem and it qualifies on an OEM's tool, switching mid-life is costly and slow (re-qual risk to the OEM's own customers). Ichor explicitly targets being the sole supplier at tool introduction and the preferred supplier for the tool's life. This is a real but narrow moat: it protects the installed base, not the next design win.
• Engineering depth + early engagement. A multi-disciplinary fluids-engineering team that embeds at customer sites and designs "within our customers' processes, design vaults, drawing standards, and part-numbering systems". This is genuinely sticky tacit knowledge — and the 10-K candidly admits the business is "largely dependent on the know-how of our employees, and we generally do not have an intellectual property position that is protected by patents". The moat is people, not patents (103 granted patents, but management says no single patent group is material).
• Bargaining power — weak over customers, moderate over suppliers. With Lam+AMAT at 76% of sales and no long-term contracts, customers hold the leverage — the 10-K lists "our customers exert a significant amount of negotiating leverage over us" as a top risk. Over suppliers Ichor has some power (multi-sourced, JIT), but the strategic answer is to vertically integrate (insource components) rather than to negotiate.
• The widening moat (thesis crux): proprietary content. Management is shifting from integrating others' parts to selling its own valves, fittings, substrates, and AFC inside the gas panel — guided from ~15% proprietary content in 2024 to ~25% in 2025, rising further. Higher own-content = higher gross margin and a deeper switching cost. This is the one place the moat is genuinely deepening, and it's the entire margin story.
• Competition: the gas/chemical-delivery subsystem market is concentrated, with Ultra Clean Technology (UCT/UCTT) the principal rival; weldment/machining is fragmented vs. numerous smaller shops. The existential competitive risk is OEM in-sourcing — if Lam or AMAT decided to build gas panels internally, Ichor's TAM collapses.

Lens 4 · Segments

Ichor reports one reportable segment. There is no clean product-line P&L in the on-disk filing markdown (the disaggregation note tables were compressed out of the ingested file), so a precise gas-vs-chemical-vs-machined revenue split is n/a from the shelf. What is sourced:

• By end-market: >90% semiconductor capital equipment; <10% aerospace/defense. A&D is the intended diversifier off a low base.
• By customer: Lam Research + Applied Materials = 76% of FY25 sales (both >10% individually).
• Revenue trend: FY23 $811.1M → FY24 $849.0M (+4.7%) → FY25 $947.7M (+11.6%). Accelerating, and accelerating hard into 2026: Q1 FY26 $256.1M (+4.7% YoY) and Q2 FY26 guided to $290–310M — management calls it "among the steepest ramps in company history," ~30% growth in two quarters.
• Cause: the WFE up-cycle (capacity for advanced logic/memory + AI/datacenter demand) plus Ichor's own content gains. The deceleration risk is the next down-cycle, not now.

---

Phase B — Measure performance

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

• Beat/tone: results came in at the upper end of guidance; revenue +15% sequentially; non-GAAP EPS +$0.15. Margin recovery driven by non-repeat of Q1-25 severance ($1.1M) and higher factory utilization.
• Guidance (the catalyst): Q2 FY26 revenue $290–310M with positive GAAP and non-GAAP EPS — the return-to-GAAP-profitability inflection.
• Balance-sheet flags: Q1 cash $89.1M (down $9.2M QoQ), driven by capex $7.1M and operating cash outflow of $(2.9)M. The OCF outflow is a working-capital build, not deterioration: AR +$22.6M and inventory +$20.5M (partially funded by AP +$27.4M) — i.e., the company is funding a demand ramp. Watch this: if revenue stalls, that WC build becomes a problem.
• The structural oddity (carries to Lens 10): Ichor posts GAAP net losses even at near-breakeven pre-tax, because income tax expense exceeds pre-tax income — $2.553M tax on $0.084M pre-tax in Q1. This is the US-valuation-allowance + Singapore/Pillar-Two cash-tax structure, not an operating problem.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the shelf (transcripts/ empty), so this is web-grounded. Tone has shifted decisively bullish over the last several quarters, tracking the cycle:

• The recurring management narrative is "proprietary content + structural margin expansion + operating leverage into 2027".
• New, more confident phrases in the Q1 FY26 call: "among the steepest ramps in company history," "unconstrained Q2 demand exceeding $300M," "all manufacturing steps for a substrate product line within the same four walls in Mexico," "advancement to full valve production later in fiscal Q2 2026".
• What they've stopped saying: the 2024–early-2025 language of "restructuring," "country exit (Scotland/Korea)," and "right-sizing." That chapter closed with FY25; the 2026 story is offense, not defense. The sentiment inflection is real and matches the financials — but it is also exactly the kind of peak-optimism tone that accompanies cycle tops.

Lens 7 · Comps

Peer set: semiconductor subsystem/component assemblers + the merchant-component names they sell into. Multiples are web-sourced with date; where a clean figure isn't sourced it's marked n/a.

Read: ICHR's TTM EV/EBITDA (~23x) sits near the semi-equipment peer median (~25x) and above its closest structural comp UCTT (~21x NTM) — despite Ichor being a thinner-margin (12% GM vs UCTT ~16%) subassembler. On forward earnings the picture is worse: ~32x FY26E EPS of $0.57, and the FY26E EBITDA the ~23x is built on is itself a cyclical-peak estimate. The multiple is not absurd for the sector at a cycle ramp, but it prices Ichor as a quality compounder, which the 12% gross margin and 76% customer concentration do not support. (One source quoted a ~139x forward P/E on a stale EPS basis and another ~32x — the spread reflects which EPS year/number is used; the ~32x on FY26E $0.57 is the defensible figure.)

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

Mostly web/price-history grounded:

• The dominant fact: ICHR is up ~424% over the trailing year and ~221% YTD 2026, from a 52-week low of $13.12 to a high of $101.91, last ~$97.70 (2026-06-22). It hit an all-time high ~$92.79 earlier in 2026. This is a WFE-cycle + AI-capex re-rating, amplified by a turnaround narrative (margins inflecting, return to profit).
• What the tape reacts to for this name: (1) WFE-cycle direction / OEM (Lam, AMAT) capex commentary — Ichor is a high-beta levered proxy on Lam+AMAT order books; (2) quarterly revenue ramp + margin prints + guidance (the Q1 FY26 guide to $290–310M was a step-change catalyst); (3) macro/AI-capex sentiment. The pattern: ICHR moves more than the SOX in both directions because it sits at the cyclical, operationally-levered, customer-concentrated end of the supply chain. It is a cycle-beta instrument, and the market treats it as one.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO — Philip (Phil) Barros (since November 2025). The signal here is strong: Barros is the engineer who built the proprietary-content franchise that is now the entire margin thesis. CTO since 2015, SVP Engineering since 2010; "led the development of the complete suite of proprietary valves, fittings, and substrate components now qualified or in final qualification with Ichor's major customers". >20 years at Ichor/Celerity; earlier an engineering project manager at Applied Materials (he knows the customer from the inside). BSME San Jose State + Stanford/Berkeley exec programs. Founder-adjacent technologist, not a financial-engineer CEO — the right archetype for a "deepen the product, insource the margin" stage.
• CFO — Greg Swyt. Entered a 10b5-1 plan Feb 26, 2026 (up to 63,273 shares).
• Skin in the game / insider activity (a flag): Both CEO Barros (up to 74,239 shares, plan dated Mar 5, 2026) and CFO Swyt adopted Rule 10b5-1 selling plans in Feb/Mar 2026 — i.e., as the stock sits near all-time highs and before the Q2 margin inflection is proven. The sizes are modest and these are scheduled/affirmative-defense plans (often tax-and-vesting driven), so it is not evidence of a problem — but it is a sentiment tell worth respecting: the people closest to the data chose to monetize into strength.
• Capital-allocation history: Mixed-to-prudent. Capex deliberately light (3.8% / 2.1% / 1.9% of sales in FY25/24/23 — a "capital-efficient, scalable" model). Deleveraged via dilution in FY24: raised $136.7M in a share issuance and used it to fully repay the $115M revolver — sensible balance-sheet repair, but it diluted holders (~33.1M → 34.4M diluted shares). Acquisitive history baked into the $335.4M goodwill (40% of total assets) — the M&A is in the past; the open question is whether that goodwill is earning its keep (Lens 10). No buybacks, no dividend.
• Net: a credible, technically-deep operator at exactly the right moment, with a clean record — modestly offset by routine-but-notable insider selling into the high.

Lens 10 · Forensic Red Flags

Forensic-analyst lens — every figure labeled.

• Goodwill + intangibles vs. equity (the biggest item). Goodwill $335.402M (unchanged YoY, no impairment) + intangibles, net $40.405M = $375.8M, ~40% of $942.9M total assets, against total shareholders' equity of $663.9M. Tangible book is only ~$288M. No impairment was taken in FY25 even through a GAAP loss year — defensible given the cyclical recovery, but goodwill is the asset most exposed if the cycle rolls and the acquired businesses underperform. Watch the annual impairment test.
• Inventory is the audit hotspot — and KPMG agrees. Inventory $231.8M (25% of total assets) carries a $37.5M excess-and-obsolete reserve, and Ichor already impaired $19.8M of inventory in FY25 (inside the restructuring charge). KPMG flagged "evaluation of excess and obsolete inventory" as the sole Critical Audit Matter. In a demand ramp this is a tailwind (write-downs reverse into utilization); in a stall it is the first place a writedown lands.
• GAAP↔non-GAAP gap is enormous (the headline forensic point). FY25 GAAP operating loss $(39.272)M vs. non-GAAP operating income $20.545M — a ~$60M bridge built from restructuring $26.6M, SBC $16.7M, amortization $8.3M, facility-shutdown (Scotland/Korea exit) $6.7M. Most of that (restructuring, country exits) is genuinely non-recurring and the FY26 GAAP/non-GAAP gap is already narrowing (Q1: $2.085M GAAP op income vs modest adjustments). But SBC of ~$16.7M/yr is a real, recurring, ~1.8%-of-sales cost that non-GAAP adds back every year — on 34.6M shares it is a persistent dilution engine, and non-GAAP EPS of $0.23 flatters the $(1.54) GAAP loss substantially. Treat non-GAAP as the cyclical truth and GAAP as the full-cost truth; the gap is legitimate but the SBC piece is permanent.
• Cash flow vs. earnings: FY25 CFO $29.9M exceeded the GAAP net loss (non-cash D&A $33.5M + inventory impairment $19.8M + SBC $16.7M bridge it) — but CFO of $29.9M did NOT cover capex of $36.2M, so FY25 free cash flow was negative ~$(6.3)M. And the multi-year CFO trend is down ($57.6M FY23 → $27.9M FY24 → $29.9M FY25) even as revenue rose — a working-capital/margin signal, not fraud. Q1 FY26 CFO was negative $(2.9)M on the ramp WC build.
• Tax red-herring resolved: the recurring "tax expense on a pre-tax loss" is the US valuation allowance (no benefit booked on US losses) + Singapore Pillar-Two cash tax (+$2.0M accrual) — structural, disclosed, not aggressive.
• Internal controls: KPMG (auditor since 2011) issued an unqualified opinion on both the financials and ICFR; management concluded disclosure controls effective as of FYE 2025 and Q1 FY26. No restatements, no error corrections flagged.

Regulatory findings (required):

• SEC Litigation Releases / AAERs: None. Verified via SEC EDGAR EFTS (LR + AAER) over 2021-06-24 → 2026-06-24.
• 10-K Item 3 (Legal Proceedings): Company states it is "presently not a party to any material litigation or regulatory proceeding". The Q1 10-Q repeats this — no material legal proceedings.
• Non-SEC enforcement (FTC/DOJ/FDA/etc.): Web search surfaced no material enforcement actions, consent decrees, or fines against Ichor Holdings. (Ichor operates ITAR-controlled A&D facilities — an export-control compliance surface to watch, but no disclosed violations.)
• Conclusion: No material regulatory or legal findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-K Item 3 as of 2026-06-24. The accounting risks are judgment risks (inventory NRV, goodwill), not enforcement risks.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (FY2026 / FY2027 / FY2028 EPS)

Built bottom-up from the latest actuals + guidance. Outputs are; inputs labeled.

Anchors: FY25 revenue $947.7M, non-GAAP op margin 2.2%, non-GAAP EPS $0.23. Q1 FY26 $256.1M; Q2 FY26 guided $290–310M (mid $300M). FY26E consensus net income ~$20.9M / EPS ~$0.57. WFE 2026 ~$135–145B, +18–20%; customers signaling 2027 growth ≥15%. Shares ~34.6M, SBC dilution ~1–2%/yr.

• Base case (cycle ramps, margins lift as guided): FY26 revenue ~$1.18B (Q1 $256M + Q2 $300M + 2H ~$310M+/qtr as ramp continues). At a non-GAAP op margin recovering toward 5–6% on operating leverage + proprietary-content mix → FY26 non-GAAP EPS ~$0.55–0.70 (consensus $0.57 anchors this). FY27 at ≥15% revenue growth ($1.35B) and margin to 7–8% as insourced valves/substrates scale → non-GAAP EPS ~$1.10–1.40. FY28 assumes the cycle is still up/plateauing ($1.45B, margin ~8–9%) → EPS ~$1.40–1.80. GAAP EPS will lag non-GAAP by the ~$16–18M annual SBC + the cash-tax drag.
• Bull case: WFE runs to $160B+, Ichor proprietary content >30%, op margin to ~9–10% by FY27 → FY27 non-GAAP EPS ~$1.60–2.00; FY28 ~$2.20+.
• Bear case: WFE 2026 disappoints or rolls in 2027 (digestion after AI-capex pull-forward); revenue flat-to-down, the 12% gross margin compresses on under-utilization, inventory writedowns resume → FY26 non-GAAP EPS ~$0.30–0.45, FY27 back toward breakeven. This is the asymmetric-downside leg — on 34.6M shares and a $3.47B cap, a cycle stall takes the multiple AND the E down together.

Brier forecast (logged conceptually — NOT executing forecast.ts create per --watchlist rules): base call = "ICHR FY26 (FYE 2026-12-25) non-GAAP EPS ≥ $0.55," p≈0.62 — supported by Q2 guidance + consensus, with the cycle as the swing factor. (Resolve at FY26 year-end.)

Lens 12 · Bull vs Bear

Bull case. Ichor is the highest-operating-leverage way to own the WFE up-cycle, plus a self-help margin story that doesn't depend on the cycle. The proprietary-content shift (15%→25%+ own valves/fittings/substrates/AFC inside the gas panel, now insourced to Mexico "within four walls") structurally lifts gross margin off the 12% subassembler floor — each point of GM on ~$1.2B is ~$12M of incremental gross profit dropping through a fixed cost base. A founder-engineer CEO who built that content is now running the company. Q2 guidance ($290–310M, +30% in two quarters, return to GAAP profit) is the proof point that the model leverages. If 2026 WFE is $135–145B and 2027 grows again, revenue + margin + dilution-light share count compound non-GAAP EPS from $0.23 (FY25) toward >$1.50 (FY27) — and a "quality subsystem compounder" re-rate is the bull's multiple.

Bear case (2–3 permanent-impairment risks). (1) Customer concentration is existential — 76% of sales from Lam + AMAT on purchase orders with no long-term commitment; either OEM in-sourcing gas panels, dual-sourcing to UCT, or simply cutting orders in a downturn permanently re-rates the franchise. (2) It is a thin-margin cyclical at a cyclical peak — 12% gross margin, GAAP-unprofitable as recently as FY25, and the AI-capex-driven WFE surge is the kind of pull-forward that digests hard; the WC build (AR +$22.6M, inv +$20.5M in Q1) becomes a cash trap if the ramp stalls. (3) The price already embodies the bull case — at ~$98 / EV-EBITDA ~23x / ~32x FY26E EPS, with the sell-side average target $81.71 sitting BELOW spot and InvestingPro pegging fair value ~$20.92, the asymmetry is poor: meeting expectations is priced; a miss reverses violently.

Pre-mortem (18 months out, thesis broke — what happened?). Most likely: WFE 2027 guidance disappoints (AI-capex digestion / China-WFE pullback / a memory air-pocket), Lam+AMAT trim orders, Ichor's utilization drops, gross margin slips back toward 10–11%, the inventory reserve grows again, non-GAAP EPS stalls near breakeven — and a $98 stock priced for 30x+ peak earnings round-trips toward the $30s–40s. The second path: a single large customer dual-sources a key gas-panel platform to UCT, capping the proprietary-content runway.

Are multiples too high? For a 12%-gross-margin, single-segment, two-customer-concentrated cyclical subassembler, yes at $98 — ~23x EV/EBITDA and ~32x forward EPS is a quality-compounder multiple on a cycle-beta business at a cycle high. The multiple is justifiable only if you believe the proprietary-content margin re-rate is permanent and the cycle runs through 2027+. Both can be true; both are also exactly what the price now requires.

Contrarian view (what the market refuses to see): The bears anchoring on a ~$21 "fair value" are using mid-cycle / trough margins and ignoring that the margin floor is structurally moving up — this is no longer the same 9%-gross-margin integrator it was; insourced content changes the through-cycle economics. The market's real blind spot cuts both ways: bulls underprice the customer-concentration tail risk, bears underprice the content-driven margin re-rate. The honest synthesis: good company, real improvement, wrong price — the edge is timing, not direction.

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

Dismantling the bull case.

• Where revenue is concentrated → the kill shot: 76% from two customers on POs, no contracts. If Lam or AMAT shifts even 10–15% of gas-panel share to UCT (the explicit alternative) or accelerates internal builds, Ichor's revenue and its proprietary-content TAM both shrink — and there's no contractual floor to catch it.
• Why the moat is weaker than bulls think: management itself says the business has "no IP position protected by patents" and is "dependent on the know-how of our employees". A people-moat walks out the door; a qualification-moat protects the installed base but not the next design win. UCT is a credible, well-capitalized rival (up 230% in 2026, targeting $3–4B revenue) competing for the same sockets.
• Most dangerous competitor bulls underestimate: Ultra Clean (UCTT) — same customers, larger scale, structurally similar, and explicitly building capacity for a $3B+ revenue base. A share-shift to UCT is the single cleanest bear catalyst.
• Worst capital-allocation / accounting flags: $335M goodwill (40% of assets) never impaired through a loss year; a $37.5M inventory reserve + $19.8M FY25 inventory writedown on a balance sheet that is 25% inventory; non-GAAP that adds back ~$16.7M of recurring SBC every year to convert a $(1.54) GAAP loss into a $0.23 "profit". The insiders (CEO + CFO) filed sale plans into the high.
• Assumptions that must hold for $98: WFE up-cycle runs through 2026 and 2027; gross margin climbs from 12% toward 15%+ on content/utilization; Lam+AMAT keep/grow their Ichor allocation; no inventory reversal. If growth disappoints 20–30% (revenue ~$0.9–0.95B vs ~$1.18B base instead), margins compress on de-leverage and non-GAAP EPS falls toward $0.20–0.35 — on which $98 is ~150x+. The valuation is a coiled spring with the cycle as the trigger.
• Single permanent-impairment scenario, and its plausibility: OEM in-sourcing of gas-delivery subsystems (the 10-K's own named risk). Plausibility low-to-moderate near-term (OEMs like the outsourced flexibility), but it is the one event that doesn't mean-revert — and it's most likely to be floated in a downturn when OEMs want to claw back margin.

Lens 14 · Management Questions (ordered by information value)

1. Proprietary content went ~15%→25% in 2025 — what is the realistic ceiling (40%? 50%?), the gross-margin uplift per 10 points of content, and what could stall the qualification pipeline at Lam/AMAT?
2. What is your honest through-cycle gross-margin target, and what utilization level does it assume — i.e., where does GM sit if 2027 WFE is flat instead of up?
3. With 76% of sales from two customers on POs, what is the real switching cost — has either customer ever dual-sourced an Ichor-incumbent gas-panel platform to UCT, and what would trigger them to?
4. Q1 working capital absorbed cash (AR +$22.6M, inv +$20.5M) on the ramp. At what revenue run-rate does the business turn FCF-positive, and what's the inventory plan if the cycle stalls mid-build?
5. The $37.5M E&O reserve and $19.8M FY25 writedown — how much of that inventory reverses into the 2026 ramp vs. is gone, and what's the demand-visibility basis?
6. $335M goodwill (40% of assets, never impaired) — what acquired-business performance underpins it, and what's the impairment-test sensitivity to a one-year revenue decline?
7. GAAP losses persist because foreign cash tax + the US valuation allowance exceed pre-tax income — what revenue/profit level releases the US valuation allowance, and what does normalized GAAP EPS look like then?
8. SBC is ~$16.7M/yr (~1.8% of sales) and central to the GAAP↔non-GAAP gap — where does SBC trend as a % of sales over the next three years?
9. The A&D segment is <10% and pitched as the diversifier — concrete 3-year revenue target, and which programs/customers underpin it?
10. AFC adoption — what share of new gas-panel design wins now spec the AFC over a merchant MFC, and what's the MFC-displacement runway with MKS/Horiba incumbents?
11. Customers are quoting $140–145B WFE for 2026 and ≥15% for 2027 — what's your own internal planning assumption, and how do you avoid over-building capacity into it?
12. Capex has been 2–4% of sales; the Mexico insourcing implies a step-up — what's the FY26–27 capex envelope and the ROIC hurdle on insourced component lines?
13. Tariffs/export controls and ITAR — quantify the exposure of your Singapore/Malaysia/Mexico footprint to a trade-policy shock and the China-WFE restriction set.
14. Capital allocation now that the revolver is repaid — buybacks (the stock is up 7x), debt paydown, or M&A, and in what priority?
15. CEO/CFO 10b5-1 plans were adopted near all-time highs ahead of the margin inflection — help shareholders read that as routine rather than a signal.

---