1. 1
   META Strong Add A / A / A− · PEG 0.89 · entry $580–630 · in zone

   Cleanest large-cap setup. 20× fwd P/E, +33% YoY rev, ad impressions +19% × price-per-ad +12% — both volume and price compound simultaneously. Capex-anxiety obscures that core monetisation is accelerating. Current $598.86 mid-zone, no earnings catalyst until late Jul — this is the lowest-event-risk Strong Add in the universe.
2. 2
   CRM Strong Add B+ / A− / A · PEG 0.86 · entry $160–205 · in zone

   13.6× fwd P/E for an enterprise SaaS leader is a value setup masquerading as software. Agentforce $800M ARR (+169% YoY) is income-statement revenue, not pipeline. 27 May print is the catalyst — Dim 1 capped at B+ until organic growth re-acceleration (FY27 H2 guide) is delivered. Asymmetric: modest Agentforce traction = rerating; disappointment = own a $146B FCF compounder at 11× EV/FCF.
3. 3
   AVGO Add (pullback) A / A− / C+ · PEG ~0.7 · entry $400–430 · ~$440 · above zone

   $73B AI backlog contractually locked across six hyperscalers (Google TPU to 2031, Anthropic 1→3GW, Meta MTIA >1GW, OpenAI 10GW). Hock Tan award triggered only at $60–120B AI revenue through FY30 — falsifiable contractual anchor. Current ~$440 has crossed above the $400–430 zone. Per quality override (Dim 1 A), a 3–5% starter position is warranted now; reload aggressively on any pullback to $400–415. 3 Jun Q2 print validates or punctures the 35–38× fwd P/E.
4. 4
   NOW Hold · Stage <$90 A / A / A− · PEG 0.83 · entry $80–95 · in zone

   A/A/A− on three dimensions — best-balanced grade card in the table. Current price ~$91 sits inside $80–95 zone. Late May Q1 print is the catalyst — target entry below $90 to establish or add. Pro Plus ACV attach + Federal renewal pacing are the two metrics that confirm the FY27 thesis.
5. 5
   TSM Add (pullback) A / A / B− · PEG 0.53 · entry $340–375 · ~9% above

   Foundry monopoly on leading-edge nodes — Apple, NVDA, AMD anchored, 59% HPC mix. +41% rev growth at PEG 0.53 is the cheapest scale-vs-multiple combination in the table. ~$410 sits ~9% above zone; per the SKILL.md quality override, Dim 1 A justifies staged starter now (3–5%) with discipline to add to target weight only on retest of $375 or post-Q2 (16 Jul) print dip.

• !AI cohort hit record highs week of 8 May. Cohort drawdowns of 15–20% historically follow record-print weeks. Stage entries over 6–8 weeks; never lump-sum at the top.
• !Memory cycle may run longer than expected — but margins revert. SNDK Q3 FY26 management confirmed NAND market "more undersupplied in Q3 than Q2," with demand above supply through CY2026. SNDK upgraded to Add (pullback); MU remains Hold. Do not add both simultaneously — combined memory exposure would exceed prudent cycle-risk limits. Cyclical margin reversion is when-not-if: NAND GMs compressed from ~65% peak to 25–35% in the 2022–23 downcycle.
• !Pre-earnings asymmetry is bad. Implied moves on NVDA (20 May), CRM (27 May), AVGO (3 Jun), MU (24 Jun), hyperscalers (late Jul) are 8–12%. Resist adding inside the 5-day pre-print window — better risk/reward post-print, off the gap.
• !Concentration is a hidden risk. Closing the active-picks gap by adding two more software names creates correlated AI-capex exposure. Spread across foundry / optics / software / power.
• !Insider clusters worth respecting. PLTR insiders sold $432.9M in 90 days incl. Thiel 1.9M shares. AAOI insider selling $29.6M / 90d. AVGO Hock Tan 200K shares sold/contributed since Dec. Treat as a hard signal, not noise — they have information you don't.
• !"In entry zone" ≠ "must buy now." The zone is permission, not obligation. If Dim 1 + Dim 2 + macro all line up, take it. If two of three are weak, wait for confirmation. Per SKILL.md quality override, Dim 1 A/A+ names above zone get a 3–5% starter — not a full skip.
• !Single-stock cap 20%. Hard rule. MU is the closest current breach risk — let it run, but the next add goes elsewhere.

• ✗Chase MU above $760. Currently ~$750, pulled back from $803 high. Consensus PT $483–533 still well below spot. Wait for the 24 Jun Q3 print; trim trigger at $850–900.
• ✗Re-enter PLTR. Position exited 14 May 2026. 88× fwd P/E competing directly with Anthropic's $1.5B enterprise-services JV (launched 4 May 2026). $432.9M insider sales 90d incl. Thiel 1.9M shares. Multiple-compression risk to ~$90 independent of fundamentals. Three risk vectors stacking, no margin of safety. Stay out; re-evaluate only at deep discount to prior $100–115 zone after 4 Aug Q2 print.
• ✗Add NVDA above $215 ahead of 20 May (7 trading days). Goldman flagged the bar is high; current $220 above $185–205 zone. Better risk/reward post-print, off any gap.
• ✗Treat NEE as an AI growth story. It's a utility with AI optionality at PEG 3.1. Add only for income and power hedge, not AI growth.
• ✗Add SNDK at market after the Q3 FY26 run-up. Q3 beat consensus by 29% ($5.95B actual vs $4.6B guide midpoint); Q4 guided $7.75–8.25B — well above prior Dec 2027 consensus of $4.19B. Verdict upgraded to Add (pullback). Wait for $1,200–1,350 zone before initiating. Stacking with MU creates correlated NAND/DRAM memory cyclicality — size ≤5% NLV on any entry.
• ✗Add AAOI before 6 Aug Q2 print. Currently ~$203, up +29% since last review — now 45% above the $120–140 entry zone. Insider selling $29.6M / 90d, Q1 slight miss ($151M vs $155M est), Texas fab ramp unvalidated. Q2 on 6 Aug is the gate; entry zone $120–140 only post-confirmation.
• ✗Deploy all capital at once. AI cohort hit record highs week of 8 May. Stage over 6–8 weeks.
• ✗Add Bloom Energy (BE) at ~$301. Never GAAP-profitable, ~135× fwd P/E, PEG 3.37. AI fuel cell thesis aspirational not demonstrated. Entry zone $180–220 (~40% below current).
• ✗Add ARM at current price ($213+). PEG 4.87 — worst in the AI-chip basket. 11.46% short interest. NVDA at 26.5× / PEG 0.35 is a strictly better AI chip bet.
• ✗Chase AMD above $420. Currently ~$436, pulling back from $460 high — approaching zone but not yet there. Meta 6GW and OpenAI MI450 commitments are real but priced in. Wait for $380–420 zone before initiating.

• ValuationIn-zone names: META (~$606/$580–630), CRM (~$179/$160–205), NOW (~$91/$80–95). PEGs 0.83–0.89. AVGO (~$440) and DDOG (~$199) have crossed above their zones — quality override permits 3–5% starters only.
• Above zoneNVDA, TSM, COHR (~$405, +16%), GEV, NEE, AMD (~$436, approaching zone), ALAB (~$223), SNDK, MU (~$750, off $803 high), LITE (~$993), BE (~$301), AVGO (~$440), DDOG (~$199), AAOI (~$203, +29%) all above entry zones. NBIS entry suspended (CapEx confirmed at Q1). Pre-earnings names (NVDA 20 May, CRM 27 May, AVGO 3 Jun) carry 8–12% implied moves.
• SentimentCohort euphoria continued through week of 8 May and beyond — optics (+4–29%), memory, custom silicon near 52W highs. Tilt slightly defensive; stage entries, don't lump sum.
• EventsNBIS Q1 done (13 May, +26% post-print, entry still suspended). NVDA 20 May, CRM 27 May, AVGO 3 Jun, MU 24 Jun, hyperscalers late Jul. Each carries 8–12% implied moves. Reserve dry powder for post-earnings down-gaps.

15% balances structural opportunity (5 in-zone Strong Add / Add candidates) against tactical caution (record-high week + clustered earnings). Below 12%: too aggressive given upcoming print risk — leaves no ammo for down-gaps. Above 18%: opportunity cost compounds at ~6% per quarter against the 26% target CAGR.

• ↓S&P drawdown >7% from 8 May high → accelerate deployment to ~10% cash.
• ↓Held name retests support post-earnings (MU $650, NVDA $195, META $580, NOW $85) → redirect dry powder into the dip.
• ↑VIX >25 sustained or two of (NVDA, CRM, AVGO, MU, META) miss → step back to 20–25% cash, wait for clarity.
• ↑Memory ASPs roll over on 24 Jun MU print → trim memory exposure, raise cash 3–5pp.