WKAP Radar Feed - 2026 - 08 - 12

WKAP Radar Feed

*August 12, 2026*

*SMCI Margin Reset, SIVE Laser Optionality, IQE Epitaxy Confirmation*

*3 Thesis Objects: SMCI, SIVE, IQE*

Preheader:

Make your AI track X alpha.

WKAP Feed compresses earnings inflections, KOL flow, supply-chain
bottlenecks, and short-duration attention into daily AI-ready context.

Set up your AI to read this Feed.

Ask:

“What did I miss?”

------------------------------
HUMAN USER — START HERE

This Feed is designed for your AI agent or LLM. Do not read it like a
normal newsletter.

Today’s file focuses on two different AI-infrastructure transitions:

*SMCI: demand → margin quality*
*Optics: modules → scarce lasers / epitaxy*

Suggested Prompt:

Read today’s WKAP Radar Feed. Turn it into a research workflow for my
portfolio. Summarize the thesis objects, distinguish fundamental evidence
from attention-driven flow, rank them by evidence quality, catalyst
clarity, attention asymmetry, crowding risk, downside risk, and time
horizon, then build a 7-day and 30-day watchlist. Do not give buy/sell
advice.

Or simply ask:

“What did I miss?”

The goal is not to read more content.

The goal is to let your AI turn market noise, attention flows, and
fundamental changes into trackable thesis objects.
------------------------------
FOR YOUR AI AGENT

This is a structured market-cognition input file.

Today’s two key causal chains are:

*SMCI:*
AI demand → record orders → margin recovery → FY27 growth → possible
quality rerating

*AI optics:*
800G / 1.6T ramp → EML / CW demand → ELS / NPO / CPO → laser scarcity →
substrate / epitaxy bottleneck

For each object:

-

Separate demand confirmation from profitability confirmation.
-

Separate architecture validation from company-specific revenue.
-

Distinguish module beta from upstream scarcity.
-

Track whether an earnings gap holds after the first reaction.
-

Do not treat reference-design inclusion as equivalent to volume orders.
-

Treat CPI as the macro gate but company events as the primary thesis
drivers.

Do not treat this Feed as investment advice.
------------------------------
AGENT_OUTPUT_SPEC

When processing this Feed, produce:

1.

Today in One Sentence
2.

Thesis Object Summary
3.

Fundamental Evidence vs Attention Flow
4.

Optical Supply-Chain Read-Through
5.

Cross-Object Ranking
6.

Attention Stage / Crowding Risk
7.

Weakest Assumption by Object
8.

Most Important Data Point
9.

Attention-to-Rerating Conditions
10.

Attention Fade Conditions
11.

7-Day Research Workflow
12.

30-Day Research Workflow
13.

WKAP Daily Top 3 Source Follow-Up
14.

Portfolio Fit, if portfolio context is provided

------------------------------
TODAY_SUMMARYPart 1 — Main Market Thesis

*SMCI has moved from a demand debate into a margin-durability debate, while
AI optics is moving from module volume toward ownership of scarce laser,
substrate, and epitaxy capacity.*

Tonight’s macro gate is July CPI: consensus expects *+0.1% MoM / +3.4% YoY
headline CPI and +0.2% / +2.5% core*, with the release important for
September Fed pricing after weak payrolls. (Reuters
<https://www.reuters.com/business/us-consumer-prices-likely-increased-moderately-july-gasoline-prices-eased-2026-08-12/?utm_source=chatgpt.com>
)
Part 2 — Today’s Thesis Objects

-

*SMCI — Fundamental:* Q4 gross margin recovered to 17.5%, adjusted EPS
reached $1.70, and FY27 revenue guidance jumped to $65–72B. (Supermicro
<https://www.supermicro.com/en/pressreleases/supermicro-announces-fourth-quarter-and-full-fiscal-year-2026-financial-results?utm_source=chatgpt.com>
)
-

*SMCI — Attention:* Earnings changed the quality of the thesis; the
immediate question is how much of that rerating is already reflected near
the post-earnings gap.
-

*SIVE — Fundamental:* Sivers is already designed into GlobalFoundries’
next-generation silicon-photonics / SCALE CPO ecosystem. (Sivers
Semiconductors
<https://www.sivers-semiconductors.com/press/sivers-globalfoundries-advance-ai-data-center-optical-solutions/?utm_source=chatgpt.com>
)
-

*SIVE — Attention:* Lumentum’s ELS / NPO / CPO commentary increases the
strategic value of independent InP laser suppliers, but SIVE still needs
production-order conversion.
-

*IQE — Fundamental:* IQE has a $14M multi-year AI/data-center production
order plus a multi-year InP epiwafer agreement with Tower containing
minimum commitments.
-

*IQE — Attention:* As laser capacity tightens, the market is moving one
step further upstream toward qualified InP epitaxy.

Part 3 — Attention Flow Today

The optical tape is shifting from *“who sells more 1.6T modules?”* toward *“who
controls the hard-to-expand optical inputs?”*

Lumentum’s earnings commentary points toward overlapping 1.6T, OCS, ELS,
NPO, and eventually CPO ramps rather than a single pluggable-product cycle.
@ParadisLabs frames the key inflection as laser supply compressing into a
larger H2’27 external-light-source ramp on top of a systems ramp already
underway. (X (formerly Twitter)
<https://x.com/ParadisLabs/status/2087317074011983904?utm_source=chatgpt.com>
)

That makes *SIVE the higher-beta architecture option* and *IQE the
better-confirmed upstream supplier*.
Part 4 — The Better Question

The key question is not:

“Which optical stock has the most CPO exposure?”

The better question is:

“Which supplier owns capacity that customers cannot quickly recreate when
1.6T, ELS, NPO and CPO all ramp at the same time?”

------------------------------
MARKET_REGIME

*RISK_TONE:* Event-driven / CPI-sensitive

*MAIN_DRIVER:* Inflation data determines the macro multiple, while earnings
and supply constraints determine stock-specific alpha.

*MARKET_CONTEXT:*

-

July headline CPI consensus: *+0.1% MoM / +3.4% YoY*. (Reuters
<https://www.reuters.com/business/us-consumer-prices-likely-increased-moderately-july-gasoline-prices-eased-2026-08-12/?utm_source=chatgpt.com>
)
-

Core CPI consensus: *+0.2% MoM / +2.5% YoY*. (Reuters
<https://www.reuters.com/business/us-consumer-prices-likely-increased-moderately-july-gasoline-prices-eased-2026-08-12/?utm_source=chatgpt.com>
)
-

A softer print would reduce pressure for renewed Fed tightening; a
hotter core print would put duration-sensitive AI names back under rate
pressure. (Reuters
<https://www.reuters.com/business/us-consumer-prices-likely-increased-moderately-july-gasoline-prices-eased-2026-08-12/?utm_source=chatgpt.com>
)
-

Stock-specific information remains unusually dense across AI hardware
and optics.

*ATTENTION_ENVIRONMENT:*

-

Earnings-driven tape.
-

Optical attention is migrating upstream.
-

Scarcity narratives are working best when backed by orders, capacity
reservations, or customer qualification.
-

Pure sympathy trades remain lower quality.

*WKAP_VIEW:*

This is a market for *fundamental reclassification*, not generic AI beta.

SMCI now needs margin persistence rather than another demand headline.

In optics, the higher-quality work is separating *real chokepoints from
thematic associations*. IQE already has contracted evidence; SIVE still has
more optionality than financial confirmation.
------------------------------
OPTICAL_CHAIN_UPDATE

Lumentum is the most important current read-through.

@ParadisLabs’ earnings-call review argues that several ramps are beginning
to overlap: 200G EML mix is accelerating; 1.6T production shipments have
begun; OCS is ramping; Lumentum has received an initial ELS order for
H2’27; and NPO appears additive rather than purely substitutive to CPO. (X
(formerly Twitter)
<https://x.com/ParadisLabs/status/2087317074011983904?utm_source=chatgpt.com>
)

The important investment implication is:

800G → 1.6T → OCS → ELS / NPO → CPO

does *not* necessarily represent five separate cycles.

It can represent multiple demand layers hitting the same laser and material
capacity.

The supply-chain map therefore becomes:

*Modules / systems:* LITE、AAOI
*Independent laser optionality:* SIVE
*InP epitaxy:* IQE
*Raw InP substrate:* AXTI

This is the core reason the upstream names may have more duration than a
traditional one-year optical-module cycle.
------------------------------
ATTENTION_TRADE_BOARDAttention Trade Board
Object Attention Stage Attention Source Why Today Hard Evidence Narrative
Gap Crowding Risk Likely Window Fade Signal
SMCI Active Earnings Margin + FY27 guide changed the thesis 17.5% GM, $1.70
EPS, $65–72B FY27 guide One quarter may overstate structural margin quality
High 1–2 weeks Margin expectation reverses
SIVE Building LITE read-through + KOL flow ELS/NPO/CPO architecture
increasingly validated GFS platform inclusion Volume orders not yet proven
Medium–High 1 month No production-order conversion
IQE Building Sector rotation + supply scarcity Upstream epitaxy becomes
more strategic as lasers ramp $14M order + Tower multi-year agreement Substrate
access can still constrain output Medium 1 month Orders fail to expandWKAP
Attention View

*SMCI* has the strongest immediate fundamental change.

*IQE* has the cleanest evidence-to-attention asymmetry in the optical
second-order trade.

*SIVE* has the largest architectural optionality but also the widest gap
between design relevance and financial proof.

Most crowded: *SMCI after earnings*.

Most likely to fade without new evidence: *SIVE*.

Best candidate for a durable optical rerating: *IQE*, if additional
customer commitments follow.
------------------------------
RADAR_OBJECT_INDEXTHESIS_OBJECT_1: SMCI

*THEME:* AI server / rack-scale infrastructure
*STATUS:* Thesis Update
*POSITION_CONTEXT:* [not provided]
*PRICE_AT_PUBLISH:* [fill at send time]
*DATE_FIRST_ADDED_TO_RADAR:* [needs update]
*SETUP_TYPE:* Earnings follow-up / Possible business reclassification
*ATTENTION_STAGE:* Active
*ATTENTION_WINDOW:* 1–2 weeks
*KEY_QUESTION:* Can gross margin remain near the mid-to-high teens while
FY27 revenue scales toward $65–72B?
------------------------------
THESIS_OBJECT_2: SIVE

*THEME:* InP lasers / external light sources / CPO-NPO
*STATUS:* Thesis Building
*POSITION_CONTEXT:* [not provided]
*PRICE_AT_PUBLISH:* [fill at send time]
*DATE_FIRST_ADDED_TO_RADAR:* [needs update]
*SETUP_TYPE:* Attention trade
*ATTENTION_STAGE:* Building
*ATTENTION_WINDOW:* 1 month
*KEY_QUESTION:* Can architectural inclusion convert into meaningful
production orders before the large 2027–2028 laser ramp?
------------------------------
THESIS_OBJECT_3: IQE

*THEME:* InP epitaxy / AI optical supply chain
*STATUS:* Confirming
*POSITION_CONTEXT:* [not provided]
*PRICE_AT_PUBLISH:* [fill at send time]
*DATE_FIRST_ADDED_TO_RADAR:* [needs update]
*SETUP_TYPE:* Possible valuation misclassification
*ATTENTION_STAGE:* Building
*ATTENTION_WINDOW:* 1 month
*KEY_QUESTION:* Can contracted demand and secured substrate supply scale
together as AI laser demand rises?
------------------------------
THESIS OBJECTSTHESIS_OBJECT_1 — SMCI

*CARD_ID:* SMCI
*CARD_TITLE:* Demand Is Proven; Margin Quality Is the New Debate
*TYPE:* Earnings Follow-up
*THEME:* AI server infrastructure
*STATUS:* Confirming
*POSITION_CONTEXT:* [not provided]
*PRICE_AT_PUBLISH:* [fill at send time]
*ATTENTION_STAGE:* Active
*ATTENTION_WINDOW:* 1–2 weeks
THESIS_SUMMARY

SMCI’s Q4 print materially changed the earnings-quality debate. Revenue
reached *$11.12B*, adjusted EPS was *$1.70 versus roughly $0.92 expected*,
gross margin recovered to *17.5%*, and FY27 revenue guidance of *$65–72B*
came well above the prior ~$52.5–53B Street expectation. (Reuters
<https://www.reuters.com/business/super-micro-forecasts-upbeat-annual-revenue-2026-08-11/?utm_source=chatgpt.com>
)

The core shift is:

*AI demand story → margin-recovery story → potential business-quality
rerating.*
WKAP_ANGLE

The surface-level frame:

“SMCI is a low-margin AI server assembler with governance baggage.”

The alternative frame:

“If enterprise mix and DCBBS can sustain mid-teens margins at much higher
revenue, the economic quality of the business has changed.”

The key research question:

“Is 17.5% gross margin repeatable?”

CORE_THESIS

SMCI generated more than $60B of new orders and entered FY27 with record
backlog, while the company now expects $65–72B of FY27 revenue. (Supermicro
<https://www.supermicro.com/en/pressreleases/supermicro-announces-fourth-quarter-and-full-fiscal-year-2026-financial-results?utm_source=chatgpt.com>
)

The important change is that gross margin recovered from *9.9% in Q3 to
17.5% in Q4*, with management attributing improvement to customer and
product mix and broader adoption of its data-center solutions. (Supermicro
<https://www.supermicro.com/en/pressreleases/supermicro-announces-fourth-quarter-and-full-fiscal-year-2026-financial-results?utm_source=chatgpt.com>
)

If that persists, earnings can rise much faster than revenue.
ATTENTION_TRADE_FRAMEAttention Source

Earnings.
Why Today

The earnings event resolved the near-term margin debate positively and
created a new question: how much follow-through should be assigned to a
fundamentally stronger FY27 setup?
Attention Stage

*Active*
Attention vs Evidence

*Hard evidence:*

-

Revenue: $11.12B.
-

Adjusted EPS: $1.70.
-

GAAP gross margin: 17.5%.
-

FY27 revenue guide: $65–72B.
-

Q4 new orders exceeded $60B. (Supermicro
<https://www.supermicro.com/en/pressreleases/supermicro-announces-fourth-quarter-and-full-fiscal-year-2026-financial-results?utm_source=chatgpt.com>
)

*Attention / interpretation:*

-

SMCI deserves a structural quality rerating.
-

The old low-margin assembler framing is no longer appropriate.
-

The next upside leg depends more on margin persistence than another
order headline.

Attention Path

Margin recovery → earnings beat → FY27 guide reset → technical acceptance →
possible quality rerating
Attention Asymmetry

The initial surprise is no longer early.

The remaining asymmetry comes from whether investors begin underwriting *17%
gross margin as sustainable rather than exceptional*.
Crowding Risk

High after the earnings gap.

The most important risk is not lack of AI demand—it is expectations moving
ahead of repeatable margin evidence.
What Could Sustain Attention

-

15%+ gross margins persist
-

FY27 orders convert
-

Enterprise customer mix expands
-

$38–40 price zone is accepted rather than immediately rejected
-

Earnings revisions continue higher

What Could Make Attention Fade

-

Margin drops back toward historical levels
-

Large orders shift between quarters
-

Customer concentration rises
-

Working-capital / financing demands increase
-

Earnings gap fails to hold

Attention-to-Thesis Conversion

If SMCI can maintain mid-teens gross margins while scaling toward its FY27
guide, this graduates from an earnings trade into a longer-duration
AI-infrastructure thesis.
WEAKEST_ASSUMPTION

The weakest assumption is that favorable Q4 customer and product mix
represents a durable margin structure rather than a particularly strong
quarter.

MOST_IMPORTANT_DATA_POINT

*Gross margin over the next two quarters.*
WKAP PATH VIEW

The quarter was genuinely above expectations.

*Short term:* WKAP sees upside toward *$38–40*, with *$40 treated as the
key call-wall / resistance zone in the current setup*.

*Long term:* the entire upgrade rests on gross margin holding. If it does,
SMCI can graduate from a post-earnings event trade into a longer-duration
holding thesis.
------------------------------
THESIS_OBJECT_2 — SIVE

*CARD_ID:* SIVE
*CARD_TITLE:* Independent Laser Optionality Gets More Relevant
*TYPE:* Attention Trade
*THEME:* InP lasers / ELS / NPO / CPO
*STATUS:* Thesis Building
*POSITION_CONTEXT:* [not provided]
*PRICE_AT_PUBLISH:* [fill at send time]
*ATTENTION_STAGE:* Building
*ATTENTION_WINDOW:* 1 month
THESIS_SUMMARY

Sivers already supplies high-power DFB lasers and arrays for AI-data-center
optical links, and its arrays are being integrated into GlobalFoundries’
silicon-photonics reference designs and SCALE CPO platform. (Sivers
Semiconductors
<https://www.sivers-semiconductors.com/press/sivers-globalfoundries-advance-ai-data-center-optical-solutions/?utm_source=chatgpt.com>
)

Lumentum’s latest read-through increases the strategic value of this
position: ELS and NPO are beginning to look like additional architectures
rather than distant CPO-only optionality. (X (formerly Twitter)
<https://x.com/ParadisLabs/status/2087317074011983904?utm_source=chatgpt.com>
)
WKAP_ANGLE

The surface-level frame:

“SIVE is a speculative future CPO supplier.”

The alternative frame:

“SIVE may become an independent laser supplier across multiple
next-generation architectures—not just one CPO implementation.”

The key research question:

“Where are the volume purchase orders?”

CORE_THESIS

The optics market increasingly needs *merchant high-power lasers* as
multiple suppliers build pluggable, NPO, external-light-source, and CPO
platforms.

Sivers’ platform inclusion makes the technical relevance credible. (Sivers
Semiconductors
<https://www.sivers-semiconductors.com/press/sivers-globalfoundries-advance-ai-data-center-optical-solutions/?utm_source=chatgpt.com>
)

But the economic thesis still requires customer qualification and
production volumes.
ATTENTION_TRADE_FRAMEAttention Source

-

LITE earnings read-through
-

CPO / NPO architecture validation
-

KOL discovery

Why Today

Lumentum receiving an initial ELS purchase order and discussing a larger
H2’27 laser ramp provides external evidence that the architecture SIVE is
positioned for is moving toward commercialization. (X (formerly Twitter)
<https://x.com/ParadisLabs/status/2087317074011983904?utm_source=chatgpt.com>
)
Attention Stage

*Building*
Attention vs Evidence

*Hard evidence:*

-

Sivers laser arrays are included in GF silicon-photonics reference
designs.
-

The arrays support GF’s SCALE CPO platform. (Sivers Semiconductors
<https://www.sivers-semiconductors.com/press/sivers-globalfoundries-advance-ai-data-center-optical-solutions/?utm_source=chatgpt.com>
)

*Attention / interpretation:*

-

Independent laser sourcing becomes structurally necessary.
-

SIVE could participate across pluggable, NPO and CPO architectures.
-

SIVE may capture the same scarcity economics as larger laser incumbents.

Attention Path

LITE ELS validation → broader independent-light-source recognition → SIVE
discovery → production-order confirmation → possible rerating
What Could Sustain Attention

-

New customer names
-

Production POs
-

1.6T-related revenue
-

CPO/NPO qualification
-

Additional hyperscaler ecosystem exposure

What Could Make Attention Fade

-

Reference designs do not convert
-

Revenue remains small
-

Financing / dilution returns
-

Larger laser suppliers absorb the demand
-

2027 ramp timing slips

Attention-to-Thesis Conversion

SIVE moves from attention trade to durable thesis when *architecture
relevance turns into committed production volume*.
WEAKEST_ASSUMPTION

The weakest assumption is that design inclusion will translate into
meaningful share of high-volume deployments.

MOST_IMPORTANT_DATA_POINT

*Signed production orders for AI optical lasers.*
------------------------------
THESIS_OBJECT_3 — IQE

*CARD_ID:* IQE
*CARD_TITLE:* Epitaxy Is Becoming the Upstream Scarcity Trade
*TYPE:* Thesis Update
*THEME:* InP epitaxy
*STATUS:* Confirming
*POSITION_CONTEXT:* [not provided]
*PRICE_AT_PUBLISH:* [fill at send time]
*ATTENTION_STAGE:* Building
*ATTENTION_WINDOW:* 1 month
THESIS_SUMMARY

IQE already has evidence that the optical-upstream thesis is entering
financials: a *multi-year $14M AI/data-center production order* plus a
*multi-year
InP epiwafer supply agreement with Tower Semiconductor*.

The Tower agreement includes a first-year minimum purchase commitment and
minimum volume commitments thereafter, supporting AI optical connectivity
including 200G/lane and next-generation technologies.
WKAP_ANGLE

The surface-level frame:

“IQE is a distant materials supplier to the optical cycle.”

The alternative frame:

“As merchant laser capacity expands, qualified epitaxy may become one of
the least replaceable layers in the stack.”

The key research question:

“Can IQE secure enough qualified substrate supply to satisfy its contracted
demand?”

CORE_THESIS

IQE sits between raw InP substrates and finished laser / photonic devices.

That position becomes more valuable if:

1.6T / ELS / NPO / CPO demand ↑ → laser output ↑ → epiwafer consumption ↑

The company already has minimum-volume commitments from Tower, making this
more than an architectural narrative.
ATTENTION_TRADE_FRAMEAttention Source

-

Optical sector rotation
-

LITE earnings
-

Upstream bottleneck recognition

Why Today

The same Lumentum commentary that validates laser scarcity also increases
the importance of qualified upstream InP epitaxy.
Attention Stage

*Building*
Attention vs Evidence

*Hard evidence:*

-

$14M multi-year AI/data-center order.
-

Multi-year Tower InP agreement.
-

First-year minimum purchase commitment.
-

Minimum-volume commitments thereafter.

*Attention / interpretation:*

-

IQE becomes a strategic non-China optical supplier.
-

Epitaxy gets scarcity pricing.
-

Multiple laser ramps create a longer-duration cycle.

Attention Path

Laser shortage → epiwafer demand → contracted capacity → earnings
conversion → upstream rerating
What Could Sustain Attention

-

Additional multi-year orders
-

Higher minimum-volume commitments
-

Secured substrate supply
-

Guidance upgrades
-

Improved EBITDA / utilization

What Could Make Attention Fade

-

Substrate constraints block growth
-

Orders remain isolated
-

Customer ramps slip
-

Epitaxy pricing fails to improve
-

Optical leaders reverse

Attention-to-Thesis Conversion

IQE already has the beginnings of a fundamental thesis.

The next upgrade comes when *multiple customers reserve capacity and
revenue growth begins converting into margin and cash generation*.
WEAKEST_ASSUMPTION

The weakest assumption is that IQE can secure enough qualified InP
substrates to fully monetize the demand arriving at the epitaxy layer.

MOST_IMPORTANT_DATA_POINT

*Follow-on contracted AI-photonics volumes plus confirmed substrate
availability.*
------------------------------
CROSS_OBJECT_ATTENTION_COMPARISONCross-Object Attention Comparison
Rank Object Attention Asymmetry Evidence Quality Catalyst Clarity Crowding
Risk Attention Window Conversion Potential
1 IQE High High Medium Medium 1 month High
2 SMCI Medium Very High High High 1–2 weeks Very High if margin holds
3 SIVE High Medium Medium Medium–High 1 month High but unprovenCleanest
Attention Trade

*IQE* — upstream scarcity is becoming better recognized while financial
evidence already exists.
Most Evidence-Backed Attention Trade

*SMCI* — earnings, margin and FY27 guidance all moved together. (Supermicro
<https://www.supermicro.com/en/pressreleases/supermicro-announces-fourth-quarter-and-full-fiscal-year-2026-financial-results?utm_source=chatgpt.com>
)
Most Crowded Attention Trade

*SMCI* after the earnings gap.
Highest Fade Risk

*SIVE* if architecture excitement does not produce purchase orders.
Best Candidate to Become a Durable Thesis

*SMCI* if margins remain structurally higher; *IQE* is the cleaner optical
candidate if contracted volumes continue expanding.
------------------------------
7_DAY_RESEARCH_WORKFLOWSMCI — 7-Day Checks

-

Track acceptance toward the $38–40 zone.
-

Monitor the $40 call-wall / resistance area.
-

Parse management comments on customer mix.
-

Update FY27 earnings estimates using 15%, 16%, and 17% GM scenarios.
-

Track backlog conversion and working capital.
-

Distinguish earnings-gap momentum from real estimate revisions.

SIVE — 7-Day Checks

-

Map every disclosed laser customer / platform.
-

Separate reference-design inclusion from volume contracts.
-

Monitor COHR and LITE commentary for merchant-laser demand.
-

Track 1.6T / ELS / NPO timing.
-

Review funding / dilution risk.
-

Identify the first plausible production-order catalyst.

IQE — 7-Day Checks

-

Review Tower minimum-volume terms.
-

Map substrate sourcing.
-

Track new optical orders.
-

Compare IQE with AXTI and Japanese substrate suppliers.
-

Monitor utilization / margin commentary.
-

Separate policy-driven attention from contracted demand.

------------------------------
30_DAY_RESEARCH_WORKFLOWSMCI — 30-Day Checks

-

Track gross-margin estimate revisions.
-

Monitor order conversion.
-

Review Q1 FY27 mix expectations.
-

Track working-capital / financing needs.
-

Test whether $38–40 becomes acceptance or rejection.
-

Upgrade to longer-duration thesis if margin durability improves.

SIVE — 30-Day Checks

-

Track production contracts.
-

Monitor next earnings.
-

Review customer qualification milestones.
-

Track CPO/NPO ecosystem announcements.
-

Compare SIVE’s content position with LITE / COHR.
-

Upgrade only when orders become economically visible.

IQE — 30-Day Checks

-

Track follow-on customer commitments.
-

Monitor InP substrate availability.
-

Review Tower volume progression.
-

Track AI/data-center revenue conversion.
-

Monitor EBITDA / cash-flow improvement.
-

Upgrade if multiple customers begin reserving capacity.

------------------------------
WKAP DAILY TOP 3

Three market sources worth feeding into today’s market chat. Not required
reading — WKAP has already extracted the signal.
1. @Anoldoldwooden — Lip-Bu Tan on Memory Architecture

URL: https://x.com/Anoldoldwooden/status/2087267721985527982

*WKAP signal:* Intel CEO Lip-Bu Tan said he historically avoided memory
because it behaved like a commodity, but now sees the sector changing as
new architectures emerge; he explicitly connected that interest with hiring
former SK hynix CEO Seok-Hee Lee and hinted that Intel is evaluating new
memory architectures. (Reddit
<https://www.reddit.com/r/intelstock/comments/1vlsb6u/new_podcast_with_lbt_dropped/?utm_source=chatgpt.com>
)

*Why it matters today:* It supports a broader WKAP thesis that AI memory is
shifting from a simple commodity cycle toward an architectural problem
involving HBM, DRAM, NAND, CXL and new memory hierarchies. A credible Intel
re-entry would also reinforce how strategic memory has become to
compute-system design.

*Themes/tickers:* INTC, SK hynix, MU, HBM, DRAM, memory architecture

*Question to ask:* “What kind of memory architecture would be strategically
differentiated enough for Intel to re-enter—and would Intel build it
internally, partner with an incumbent, or integrate memory more tightly
with its CPU / packaging stack?”
2. @SemiAnalysis_ — Google TPUv8i Software Bring-Up

URL: https://x.com/SemiAnalysis_/status/2087298210066829389

*WKAP signal:* SemiAnalysis reports that Google’s next-generation *TPUv8i
is already undergoing internal software bring-up on the g3 codebase*,
suggesting Google is advancing the software stack before broad hardware
deployment. (X (formerly Twitter)
<https://x.com/SemiAnalysis_/status/2087298210066829389?utm_source=chatgpt.com>
)

*Why it matters today:* Custom AI accelerators are becoming an integrated
hardware-software product rather than a chip-only race. Earlier software
enablement can shorten deployment cycles, improve utilization and
strengthen Google Cloud’s ability to monetize proprietary compute.

*Themes/tickers:* GOOG, TPUv8i, custom ASIC, AI accelerators, GCP, software
stack

*Question to ask:* “Does early TPUv8i software bring-up imply a faster
hardware deployment cycle—and which networking, memory, packaging and
optical suppliers must ramp before the accelerator reaches production
scale?”
3. @ParadisLabs — Lumentum Earnings and the Laser Bottleneck

URL: https://x.com/ParadisLabs/status/2087317074011983904

*WKAP signal:* ParadisLabs frames Lumentum’s quarter as evidence that *1.6T,
OCS, ELS, NPO and CPO are stacking into overlapping optical demand waves*,
with the large external-laser ramp moving toward H2’27 rather than
replacing the current transceiver cycle. (X (formerly Twitter)
<https://x.com/ParadisLabs/status/2087317074011983904?utm_source=chatgpt.com>
)

*Why it matters today:* This is the most important read-through for SIVE
and IQE. If multiple architectures hit the same laser supply chain
simultaneously, the investable bottleneck shifts upstream from module
assembly toward *high-power InP lasers, substrates and epitaxy*.

*Themes/tickers:* LITE, AAOI, SIVE, IQE, AXTI, 1.6T, ELS, NPO, CPO, InP

*Question to ask:* “If the H2’27 ELS / CPO ramp lands on top of an already
strong 1.6T cycle, which suppliers have enough qualified InP capacity to
capture the incremental demand without sacrificing margins?”