PREVIEW — Synthos Deep Dive v2 template · not published · founder review · 2026-07-03
Sanmina SANM
Technology · Hardware, Equipment & Parts · Synthos Deep Dive · 2026-07-03
$217.61
Watch
Risk 7Growth 5Exponential 3Fair value $200 $130–$285
The 20-second read
What it does
Sanmina Corporation (Nasdaq: SANM) is a global electronics manufacturing services (EMS) provider: components, finished goods, repair, supply-chain management, and post-sale support for original equipment manufacturers. Two segments: Integrated Manufacturing Solutions (IMS) — PCB assembly through full-system build and logistics — and Components, Products and Services (CPS) — interconnect systems …
Where it stands
$217.61 · Watch · fair value ~$200 (-8% vs price) · Risk 7/10, Growth 5/10
Where it's going
SANM's AI-rack ramp is real but the stock already trades above the Street's lone $200 target with the CEO selling and momentum rolling over — it gets interesting near ~$175 (≈13–14× FY27E EPS); a Q3 sequential revenue decline worse than the ~$3.4B estimate breaks the story.
The AI-rack revenue step-up proving lumpy: consensus already models growth collapsing from +74% to +2% in two years, and the multiple has re-rated for the good year
One-line thesis. Sanmina — a 46-year-old, thin-margin electronics contract manufacturer — just caught the AI-infrastructure wave: quarterly revenue doubled year-over-year in Q2 FY26 ($4.01B vs $1.98B) and FY26 consensus revenue is +74%. But the estimates themselves say this is a step-function, not a compounding curve (+15% FY27E, +2% FY28E, FY28E EPS below FY27E), the stock already sits ~9% above the Street's only price target, the CEO sold ~$27M of stock in May, and price is breaking down below its 50-DMA — so the honest call is Watch, not chase.
◆ Synthos call — WatchSANM's AI-rack ramp is real but the stock already trades above the Street's lone $200 target with the CEO selling and momentum rolling over — it gets interesting near ~$175 (≈13–14× FY27E EPS); a Q3 sequential revenue decline worse than the ~$3.4B estimate breaks the story.
Downside Risk (lower = safer)
7/10 · High
Beta 1.56, 2.3% net margins, leverage added post-FY25 (net-debt/EBITDA 1.05× TTM vs net cash at FY-end), price below the 50-DMA with negative MACD, and a cluster of insider sales incl. the CEO's ~$27M — thin coverage (one price target) adds blind-spot risk.
Growth Quality
5/10 · Moderate
Revenue +74% FY26E is a step-function (acquisition/AI-rack driven), fading to +15% FY27E and +2% FY28E; gross margin 8.5%, ROIC 7.4% — but cash conversion is genuinely strong (income quality 3.56, FCF yield 6.3%).
Exponential Potential
3/10 · Low
The second derivative is sharply negative — +74% → +15% → +2%, with FY28E EPS ($12.29) below FY27E ($12.98); an 8.5%-gross-margin contract manufacturer rides the AI build-out as pass-through volume, not pricing power. A step-up, not an exponential.
⚖ Reverse-DCF cross-checkMarket-implied growth ≈ 28%/yrTo justify today’s $218, earnings would have to compound roughly 28% a year for 10 years (9% discount rate). Analysts forecast ~17%/yr, so the market is pricing in MORE than what the Street expects.What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
Buy — CoreOwn it as a foundation — start or add now, size it for years, let dips be gifts.
Buy — TacticalGood price + confirmed trend + a defined exit — buy the setup, not a marriage.
WatchWe want the business, just not at this price/setup — act only when the listed trigger hits.
HoldFine to keep if you own it — no reason to buy more; new money does better elsewhere.
AvoidDon't own it — the problem is the business or the expectations, so a cheaper price won't fix it.
In plain English
Sanmina builds electronics for other companies — circuit boards, enclosures, and full systems for customers in cloud computing, defense, healthcare, telecom, and industrial markets. It's the factory behind other people's brands. That business is high-volume and very low margin: for every $100 of product, Sanmina keeps about $8.50 before overhead and roughly $2.30 as profit.
What changed: Sanmina plugged itself into the AI data-center build-out (it bought a large data-center systems manufacturing business — see §1), and revenue exploded — the most recent quarter was double the same quarter a year ago. The stock more than doubled in twelve months on that news.
Here's the catch, in everyday terms:
The growth spike looks like a one-time step up, not a rocket. Analysts expect +74% revenue growth this fiscal year, then +15%, then just +2% — and earnings per share are actually expected to dip in fiscal 2028.
The stock already trades above what the Street thinks it's worth (the only published target is $200; the stock is ~$218).
The people who know the company best are selling — the CEO sold about $27 million of stock in May, and the CFO and other officers sold too. No insiders bought.
The chart has rolled over — the price has dropped 23% from its high and fallen below its 50-day trend line.
Our three scores: Downside Risk 7/10 (fairly high) — razor-thin margins, new debt, weak momentum, insider selling. Growth Quality 5/10 (average) — huge but likely temporary growth; the real bright spot is strong cash generation. Exponential Potential 3/10 (low) — growth is decelerating hard, the opposite of what we look for in an exponential.
The one big worry: contract manufacturing for AI data centers is lumpy, pass-through business. If the rack-building wave pauses — or the anchor customer relationship shifts — a stock that re-rated for one great year gives it back fast.
Solid = price · dashed = 50-day average · dotted = 200-day average · amber = 52-week high/low. Price above both averages is an uptrend.
Bollinger Bands 20-day average ± 2 standard deviations
The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.
Blue crossing above amber (bars flip green) = momentum turning up; below (bars red) = turning down. Bar height = the size of that gap.
Relative performance vs S&P 500 & its sector (XLK (sector)), set to 100 a year ago
Solid = SANM · dashed = S&P 500 · dotted = XLK (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.
Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.
Key stats an RIA wants
Price$217.61
Market cap$12B
P/E trailing45×
P/E FY26E / FY27E19× / 17×
EV / Sales1.1×
EV / EBITDA21.5×
Gross margin8.5%
Net margin2.3%
Dividend yield0.00%
Beta1.561
52-wk range$96 – $283
RSI(14)34
50 / 200-DMA$240 / $169
12-mo return+115% (SPY +21%)
Street target$200 ($200–$200)
Analyst grades5 Buy · 10 Hold · 2 Sell
FMP ratingB-
Next earnings2026-08-05
What the experts actually said 0 traceable claims on SANM · showing the highest-conviction voices
Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.
1. What it is
Sanmina Corporation (Nasdaq: SANM) is a global electronics manufacturing services (EMS) provider: components, finished goods, repair, supply-chain management, and post-sale support for original equipment manufacturers. Two segments: Integrated Manufacturing Solutions (IMS) — PCB assembly through full-system build and logistics — and Components, Products and Services (CPS) — interconnect systems (printed circuit boards, backplanes, cable assemblies), mechanical systems (enclosures, precision machining), plus memory, storage, RF, optics, microelectronics, defense/aerospace products and cloud-enabled manufacturing software. End markets: industrial, healthcare, defense & aviation, automotive, telecom, and cloud computing. Founded 1980, IPO 1993, headquartered in San Jose, CA; founder Jure Sola is Chairman & CEO; ~32,000 employees. Fiscal year ends late September/early October (FY25 = year ended 2025-09-27).
Revenue mix (FY25, from filings):
By segment: IMS $6.51B (80%) · CPS third-party $1.62B (20%).
By geography: Americas $4.65B (57%) · Asia-Pacific $2.61B (32%) · EMEA $0.87B (11%). A majority-Americas footprint is a genuine differentiator versus Asia-centric EMS peers in a tariff/onshoring world.
What just changed (context from the public record, not the data file): Sanmina agreed in May 2025 to acquire ZT Systems' data-center infrastructure manufacturing business from AMD, making it a rack-scale manufacturing partner for AI systems. The data file shows the fingerprints even without naming it: Q1 FY26 revenue +59% YoY ($3.19B), Q2 FY26 +102% YoY ($4.01B), deferred revenue of $878M on the FY25 balance sheet (customer advances), and quarterly interest expense jumping from ~$5M to $32M (deal financing). Everything numeric in this note comes from the data file; the deal name is the one labeled exception.
2. The expert thesis (traceable)
No expert-panel coverage — this note is fundamentals-driven. The Synthos KB contains zero SANM claims from zero voices. SANM entered the research queue via the quant momentum screen (+115% twelve-month return), not via conviction voices. Per house standard, that means: no borrowed conviction, no synthetic quotes — the bull and bear cases in §3 are built entirely from the filed financials, consensus estimates, insider filings, and technicals in the data file. Treat the conviction rating (None) accordingly: this is a lower-confidence note than a KB-backed name, and it is scored and sized that way.
3. Synthos scores & the Bull / Base / Bear cases
Three scores, 0–10, each anchored to real metrics (not probabilities we can't honestly calibrate):
Score
0–10
The read
Downside Risk(lower = safer)
7 · High
Beta 1.56; net margin 2.3% (no cushion when volume wobbles); net cash at FY25-end flipped to net-debt/EBITDA 1.05× TTM as deal financing landed; price below the 50-DMA with MACD negative and a −23% drawdown underway; CEO sold ~$27M of stock in May with no offsetting buys; only one published price target — thin coverage means thin early warning. Offsets: current ratio 1.71, FCF yield 6.3%, FMP rating B− (overall 3/5, but debt-to-equity scores 1/5).
Growth Quality
5 · Average
FY26E revenue +74% is real but acquisition/ramp-driven, and consensus decays it to +15% (FY27E) → +2% (FY28E) with FY28E EPS below FY27E. Gross margin 8.5%, ROIC 7.4%, R&D 0.3% of sales — structurally commodity economics. The genuine strengths: income quality 3.56 (cash flow far exceeds GAAP income), FY25 FCF $473M, capex only 2.1% of revenue.
Exponential Potential
3 · Low
The second derivative is sharply negative — the defining anti-exponential signature. An 8.5%-gross-margin pass-through manufacturer captures AI volume, not AI economics.
The three cases (our own scenario model — assumptions labeled; each target is a ~12–18-month fair value). We deliberately do not attach probabilities: the base case is the expected path; the cases bound the range.
Case
Key assumptions
Fair value
Bull
AI-rack demand keeps beating (Q2 FY26 beat revenue estimates by 22%); FY27E EPS lands ~$13.9 (consensus high) and the market pays ~20× for a re-rated US-based AI-infrastructure manufacturer. Roughly a retest of the 52-wk high ($282.72).
~$285 (+31%)
Base(our anchor)
Estimates roughly hit — FY27E EPS ~$12.98 — but a thin-margin EMS with growth fading to ~2% earns a mid-teens multiple: ~15.5× FY27E. This lands on the Street's $200 target — labeled: our multiple assumption, their number.
~$200 (−8%)
Bear
The step-up proves lumpy: Q3's guided sequential decline (~$3.40B vs $4.01B) extends, FY27E EPS misses toward the ~$11.7 consensus low or below (~$11.5), and the multiple de-rates to ~11× as "AI infrastructure" reverts to "contract manufacturer."
~$130 (−40%)
Synthos fair value = the base case, ~$200 (−8%). Note what the shape of the range says: the bear (−40%) is bigger than the bull (+31%) from here — negative skew at the current price. That asymmetry, not the growth headline, is why the verdict is Watch. This is a tracked call — the Forecaster Scorecard grades it once it matures.
4. Exponential Potential
Synthos separates compounders from exponentials (accelerating, multi-baggers-from-here). SANM is neither — it is a step-function:
Forward growth: FY26E revenue $14.12B (+74%) off FY25's $8.13B — a huge one-year jump. FY26E EPS $11.16 vs FY25's ~$5.95 estimate-basis (+88%).
Acceleration (the 2nd derivative) is negative: +74% (FY26E) → +15% (FY27E, $16.24B) → +2% (FY28E, $16.54B). EPS follows: $11.16 → $12.98 → $12.29 (down 5%). The same consensus that gives the growth takes away the durability — analysts are modeling a plateau, not a curve. (Caveat: FY28 rests on one analyst; FY26/FY27 on three.)
Economics cap the dream: gross margin 8.5%, EBITDA margin 5.0%, R&D 0.3% of revenue. Volume scales; value capture doesn't. The AI build-out flows through Sanmina rather than compounding inside it.
Room to run: at $11.7B market cap there is nominal room, but 1.08× EV/S on ~2% terminal growth is not the setup for a multibagger — the FY26 step-up is largely in the price.
Exponential Potential: Low (3/10). The near-opposite of the profile we hunt: growth is fast but decelerating, and the business keeps pennies of each incremental AI dollar.
Revenue: FY25 $8.13B, +7.4% (FY24 $7.57B, −15.3%; FY23 $8.94B, +12.8%; FY22 $7.92B; FY21 $6.74B; FY20 $6.95B). Read that history honestly: five years of cyclical chop around $7–9B — this was a no-growth business until the current ramp.
Quarterly trajectory (the step-up): Q3 FY25 $2.04B → Q4 $2.10B → Q1 FY26 $3.19B (+59% YoY) → Q2 FY26 $4.01B (+102% YoY). H1 FY26 revenue ($7.20B) nearly equals all of FY25. But Q3 FY26 is estimated at ~$3.40B — a guided sequential decline of ~15% — confirming lumpiness.
Margins: FY25 gross 8.8%, operating 4.4%, net 3.0% (GAAP net income $245.9M, diluted EPS $4.46). TTM: gross 8.5%, EBITDA 5.0%, net 2.3%. Q2 FY26 GAAP diluted EPS was $1.70 vs non-GAAP $3.16 — the GAAP/non-GAAP gap blew out post-deal (amortization/acquisition charges); underwrite with that gap in view.
Cash flow (the bright spot): FY25 operating CF $620.7M, capex −$147.4M (2.1% of revenue... light), FCF $473.3M. Income quality TTM 3.56 — cash income far exceeds GAAP income. Honest caveat: FY25 OCF was boosted by +$738M of "other working capital" (largely the $878M deferred-revenue/customer-advance build) while inventory consumed −$543M — some of that cash is customers pre-funding the ramp, i.e., timing, not margin.
Balance sheet: at FY25-end (2025-09-27): cash $966M, total debt $394M → net cash $572M, current ratio strong. Since then the TTM metrics show net-debt/EBITDA 1.05× (EV $12.26B vs $11.66B market cap ⇒ ~$0.6B net debt) and quarterly interest expense stepped from ~$5M to $32.1M (Q2 FY26) — deal financing landed after the last filed annual balance sheet. The current debt figure is not in the data file — flagged as a gap, not guessed.
Capital returns: no dividend (yield 0%); buybacks $113.8M FY25 ($254.3M FY24, $331.1M FY22). Share count down from 69.0M (FY20) to 54.2M — a decade-long cannibal, though repurchases slowed as the deal took priority.
6. Valuation — priced in or room?
Two honest lenses, and they disagree:
Trailing GAAP says expensive:45.2× TTM EPS, 4.9× book, 21.5× EV/EBITDA — all inflated by the GAAP/non-GAAP gap and the half-completed ramp. FMP letter rating B− (P/E score 2/5, D/E score 1/5).
Forward non-GAAP says merely full:19.5× FY26E ($11.16) → 16.8× FY27E ($12.98) → 17.7× FY28E ($12.29) — note the forward multiple rises into FY28 because EPS is modeled to fall. EV/S 1.08× and FCF yield 6.3% are the value-ish anchors; forward PEG 2.78 says growth is already paid for.
The tell: for a business whose consensus terminal growth is ~2%, ~17× is not a bargain — it embeds continued AI-rack momentum. And the stock trades ~9% above the Street's only published target ($200; high = low = median — effectively one analyst's view, context not gospel) with a 5 Buy / 10 Hold / 2 Sell ratings skew.
Net read: modestly overvalued. You are paying a fair-to-full multiple for the best year in the company's history at the exact moment consensus models the growth stalling. The entry that changes the math is ~$175 (≈13.5× FY27E) — close to our bear-bull midpoint and a level where the 6%+ FCF yield does the underwriting.
7. Technicals (from the tech block)
Trend: long lens still up, short lens broken. $217.61 is −9.2% below the 50-DMA ($239.75) but +28.6% above the 200-DMA ($169.19). MACD −2.33 (negative and confirming the breakdown).
Location:−23.0% off the 52-week high ($282.72) — also the max drawdown from peak, i.e., this correction is the deepest of the past year — and +127% off the 52-week low ($95.79).
Momentum: RSI(14) 33.9 — weak, approaching (not yet at) the oversold line. This is a falling knife's neighborhood, not a base.
Relative strength: +115% 12-mo vs SPY +21% / QQQ +31%; +67% 3-mo vs SPY +15% / QQQ +24%. Enormous outperformance — the screen caught it late in the move, which is exactly when momentum names are most fragile.
Read:distribution, not accumulation. A parabolic 12-month run has cracked below its 50-DMA on negative MACD while insiders sell. For a Watch name that's convenient: let it either reclaim the 50-DMA on a strong Q3 print or come to us near the high-$160s/$170s where the rising 200-DMA and the valuation math converge.
8. Moat & competitive position
EMS is a scale-and-switching-cost business with structurally thin economics. Sanmina's genuine edges: (1) a majority-Americas manufacturing footprint (57% of FY25 revenue) at a moment when tariffs and onshoring push OEMs toward US/Mexico capacity; (2) vertical integration via CPS — PCBs, backplanes, enclosures, optics — that most assembly-only rivals lack; (3) regulated-market qualification (defense/aero, medical) where switching is slow; (4) the new rack-scale AI systems capability, which ties it to the largest capex cycle in tech. Against that: 8.5% gross margin and 7.4% ROIC say pricing power is minimal; customers are giant OEMs/hyperscalers with all the leverage; and the big EMS comps (Jabil, Celestica, Flex, Foxconn) fight for the same AI-rack sockets.
Peer set (FMP-supplied, market cap): a poor-fit list — Badger Meter $4.4B, Cognex $11.6B, Descartes $6.3B, ESCO $8.9B, Littelfuse $10.7B, MKS $24.9B, Universal Display $3.8B, Paylocity $6.1B, TTM Technologies $15.5B (the only true adjacent, PCB-centric), Vontier $4.1B. The relevant comparators — Celestica, Jabil, Flex — are not in the supplied set; judge SANM against the EMS/AI-infrastructure cohort, not this heterogeneous list. Data caveat, flagged per house standard.
9. Management, capital allocation & guidance
Founder-led: Jure Sola co-founded Sanmina in 1980 and remains Chairman & CEO — decades of cycle experience, and the ZT-scale bet shows continued aggression.
Capital allocation: historically shareholder-friendly on the buyback (share count −21% since FY20; $113.8M repurchased FY25), no dividend, light capex (2.1% of revenue), and — until the deal — a net-cash balance sheet. The pivot to debt-funded M&A is the biggest allocation decision in company history; the data file can't yet score its returns.
Insider activity (the yellow flag): the recent Form 4 tape is all sales, no buys: CEO Jure Sola sold 118,368 shares at $228.84 on 2026-05-06 (~$27M, ~9% of his reported 1.23M-share stake); CFO Jonathan Faust sold 2×5,038 shares at ~$265.73–265.86 (2026-05-29); EVP HR Alan Reid sold in four lots ($218–254, May 2026); director David Hedley sold 500 at $219.52. Some of this is likely 10b5-1 diversification after a +115% year — but a CEO trimming ~$27M into a parabolic move, with zero insider buys on file, is a tell we weight.
Guidance/estimates: the company has beaten EPS estimates five straight quarters (most recently $3.16 vs $2.42, a 31% beat, with revenue $4.01B vs $3.28B expected). Earnings-call guidance detail is not in our data file — flagged as a gap rather than paraphrased from memory.
10. Catalysts & what to watch
Next earnings: 2026-07-27 (Q3 FY26; Street EPS $2.78, revenue ~$3.40B). The single most important line: does revenue really step down ~15% sequentially from Q2's $4.01B, and does management frame it as timing (program transitions, customer-advance recognition) or demand? A beat-and-raise above ~$4B changes this note's tilt; an in-line-or-worse print validates the lumpiness thesis.
AI-rack program news: new rack-scale wins or expansions (or any sign the anchor relationship narrows) — the swing factor behind the entire step-up.
Balance-sheet disclosure: the first full post-deal balance sheet — actual debt, integration costs, and whether the deferred-revenue cushion ($878M at FY25-end) persists or unwinds.
Onshoring/tariff policy: a majority-Americas EMS is a policy beneficiary; incremental reshoring announcements help.
Estimate revisions & coverage: FY28 rests on one analyst and the Street shows one price target — new initiations could move the reference frame materially in either direction.
Thesis tripwires (what would change the call): Q3 FY26 revenue materially below ~$3.4B or Q4 guided down again (bear tilt); two consecutive quarters back above $4B with gross margin holding ≥8.8% (bull tilt); insider buying by Sola or Faust (bull tilt); net-debt/EBITDA rising above ~1.5× (bear tilt).
11. Key risks
Growth-cliff risk (the dominant one): consensus itself models +74% → +15% → +2%, with FY28E EPS below FY27E. If the AI-rack ramp is a project, not a franchise, the stock re-rates from "AI infrastructure" back to "EMS" — historically a low-teens-multiple, 1–3%-net-margin category.
Customer concentration / pass-through economics: rack-scale AI manufacturing hangs on a small number of giant customers with total pricing leverage over an 8.5%-gross-margin supplier. (Per-customer concentration figures are not in the data file — flagged, not guessed.)
Integration & leverage: the largest deal in company history, financed with debt (interest expense ~6× the prior run-rate); execution slippage hits a 2.3%-net-margin P&L hard.
Working-capital cycle: inventory $1.99B at FY25-end (up $545M YoY), cash-conversion cycle 105 days; FY25's cash flow leaned on $738M of other-working-capital inflows (customer advances) that can reverse.
Insider selling: all-sale, no-buy Form 4 tape including the CEO's ~$27M — modest as a percentage of holdings, but directionally one-way.
Momentum unwind: beta 1.56, −23% drawdown in progress, below the 50-DMA on RSI 34 — momentum screens surface names late; the exit door is narrow for a mid-cap.
GAAP/non-GAAP gap: Q2 FY26 GAAP EPS $1.70 vs non-GAAP $3.16 — the forward multiples in §6 rest on the adjusted series; the statutory earnings are much smaller.
Cyclicality (the base rate): FY20–FY25 revenue chopped between $6.7B and $8.9B with a −15% year in FY24 — this company's own history argues against extrapolating any single great year.
12. Verdict, position sizing & monitoring
Watch. Sanmina is a better business than its stock chart history suggests — founder-led, cash-generative (6.3% FCF yield), majority-Americas, and now genuinely attached to the AI data-center build-out with revenue that doubled year-over-year. But the honest ledger tips the other way at $217.61: the stock trades above the Street's lone $200 target and our own ~$200 base case; consensus models the growth collapsing to +2% within two fiscal years with FY28E EPS down; the CEO and CFO are sellers; the chart has broken its 50-DMA; and there is no expert-panel conviction behind the name. The risk/reward is negatively skewed (−40% bear vs +31% bull). That is a Watch, not a Buy — and not a short either, given the beat streak and cash flow.
Sizing:no position today. The setup that would justify a tactical entry: ~$175 (≈13.5× FY27E EPS, near the rising 200-DMA) or a Q3 print on 2026-07-27 that beats the ~$3.40B revenue estimate and reclaims the 50-DMA — then ≤1–2% of the flagship, sized as a cyclical, never core.
Monitoring: re-underwrite on the §10 tripwires; formal re-score at Q3 FY26 earnings (2026-07-27). This verdict is logged as a tracked Synthos call as of 2026-07-06 at $217.61.
Single biggest risk: the AI-rack step-up proving lumpy/one-time — a thin-margin manufacturer priced off its best year, with the deceleration already written into the very estimates that support the multiple.
Provenance & disclosures
Traceability:0 KB claims, 0 voices — no expert-panel coverage; per house standard this is a fundamentals-driven note surfaced by the quant momentum screen. No conviction is borrowed or synthesized; kb_net_conviction is null because there is nothing to aggregate.
Data as-of: fundamentals 2026-03-28 (Q2 FY26 10-Q) · estimates & prices 2026-07-06 (FMP pull in the data file) · insider filings through 2026-06-01. Forward figures are analyst consensus (FMP), labeled as estimates; FY28 rests on one analyst and FY26/FY27 on three — treat with corresponding humility.
Acquisition caveat: the ZT Systems / AMD deal is named from the public record — it is the one fact in this note not sourced from the data file; every number is from the file. The post-deal debt balance is not in the file (annual balance sheet ends 2025-09-27) and is inferred only directionally from TTM EV and interest expense — flagged, not fabricated.
Earnings-quality caveat: GAAP TTM EPS ($4.79) sits far below the non-GAAP series the Street quotes (FY26E $11.16); FY25 operating cash flow was flattered by ~$738M of working-capital/customer-advance inflows. Underwrite both gaps.
Street-coverage caveat: the price-target block contains a single $200 target (high = low = median) and 17 ratings skewed to Hold — consensus here is thin context, not a robust anchor.
Peer caveat: the FMP-supplied peer list omits the relevant EMS comps (Celestica, Jabil, Flex); judge against that cohort.
Not investment advice. Independent research, educational and informational only, never personalized. Hypothetical/forward figures are labeled; the only performance numbers Synthos will headline are the live, real-money Flagship's.
Version: 2026-07-06. Prior versions available via the deep-dive version dropdown ("based on the info at the time").