Broadcom reports on Wednesday and the August employment report lands on Friday, and between them they frame the week. $AVGO’s quarter to 2 August arrives after the US close on the 2nd against a guide of $16.0bn in AI semiconductor revenue, more than 200% above a year ago – but the number that moved the stock last time was the one management left alone. Contemporaneous coverage read the decision to reiterate rather than raise a fiscal 2027 AI target of “in excess of $100 billion” as the trigger for what followed: the shares fell 12.6% in the session after the June print and a further 7.9% the day after that, and they have not made it back – the stock still sits about 25% below a 52-week high of $495.00. It reports into a bar Nvidia raised exactly a week earlier, with $96.2bn of July-quarter revenue against a $91.0bn guide, an October-quarter guide of $108bn and management’s forecast of about 70% revenue growth in fiscal 2028. Tuesday is the crowded day. Partners Group publishes interim figures at 7am Central European time, before the European open; Palo Alto Networks, Dell, MongoDB, Credo and Couche-Tard all follow after the North American close. Partners Group and Couche-Tard sit on opposite sides of the same transaction – Partners Group is among the holders that have irrevocably committed their Żabka shares to Couche-Tard’s tender offer. Monday opens in Hong Kong, where Z.AI’s board meets to approve the first interim results since a January listing billed as the market’s first for a pure-play large-language-model developer. Thursday closes on Lululemon, five days before a new chief executive starts, at about eleven times the low end of the company’s own guide. And Friday brings the August employment report at 8:30am Eastern, into a market that spent last Friday repricing after Kevin Warsh’s Jackson Hole remarks flagged worry about inflation; July’s 23,000 decline left the unemployment rate at 4.1%, and forecasts for August are unusually dispersed – Trading Economics has the consensus near 12,000, Capital Economics at 90,000 – which is its own comment on how hard this labour market is to read. The S&P 500 still finished the week 0.5% higher at 7,711.76, roughly 1.1% below its 13 August record close of 7,798.99. Sixteen names make our calendar; Hewlett Packard Enterprise, NetApp, Ciena, Zscaler, Samsara and Guidewire fill in the rest.
🇺🇸 Broadcom
$AVGO reports fiscal third-quarter results Wednesday after the US close – the print the AI-infrastructure complex is waiting on, and the one with the widest read-across in the week. The shares closed Friday at $368.79, down 0.74%, having risen 4.5% on Thursday in the wake of Nvidia’s numbers; the 52-week range is $287.17 to $495.00, which leaves the stock about 25% below its high, and on a closing basis the peak was $481.57 on 2 June – the day before the last set of results. Consensus for the quarter sits on the guide at roughly $29.2bn of revenue and $3.22 of non-GAAP earnings across more than twenty estimates. The valuation is the part worth holding in mind: on those same estimates the stock trades near 32 times this fiscal year’s earnings and closer to 19 times next year’s, which is another way of saying the fiscal 2027 number is doing most of the work.
Fiscal Q2, the quarter ended 3 May, set the base. Revenue was $22.19bn, up 48%, split $15.01bn of semiconductor solutions, up 79%, and $7.18bn of infrastructure software, up 9%. AI semiconductor revenue was $10.8bn, up 143% and slightly ahead of the $10.7bn guided three months earlier. Adjusted EBITDA was $15.24bn, or 69% of revenue; non-GAAP diluted earnings were $2.44 and free cash flow $10.26bn, 46% of revenue. Guidance for this quarter is about $29.4bn of revenue, up 84%, with non-GAAP operating income near 67% of revenue and adjusted EBITDA near 68%, and $16.0bn of that from AI. The arithmetic behind the full year is the more interesting part. AI revenue was $8.4bn in the January quarter and $10.8bn in April, so the first three quarters against the $16.0bn guide come to about $35.2bn – against a full-year AI figure the company has put at $56bn. That implies roughly $20.8bn in the October quarter, about 30% more than this quarter’s guide, before fiscal 2027 is asked to grow by at least about 79% again – which is the step from $56bn to “in excess of $100 billion”, not another doubling.
The demand disclosure is unusually specific and unusually incomplete at the same time. Bookings for AI semiconductors were more than $30bn in the April quarter against the $10.8bn shipped, and Hock Tan told the call that visibility now runs “all the way to 2028”. Six customers carry the bulk of it: Google, Anthropic, OpenAI and Meta are named, with two unnamed buyers accounting for about $6bn of orders that begin shipping late in 2026. What the company has not restated is a single backlog total. The roughly $73bn most often quoted was given in December for the following eighteen months and has not been updated since; the bookings and shipment figures indicate the backlog grew during the quarter, but Broadcom has not disclosed a current total. Mix is moving as well: networking was almost 40% of AI revenue in the April quarter and management expects that to drift back toward 30% as custom accelerators ramp. At group level, gross margin was 77.1%, down 230 basis points, and is guided to about 74% this quarter – a mix effect as semiconductors grow faster than infrastructure software, which the chief financial officer said does not represent a structural change in semiconductor margin.
Our readThe fiscal 2027 AI framing is the swing factor, not the July quarter. The quarter itself is guided tightly and consensus sits on it; what is unresolved is whether the “in excess of $100 billion” formulation gets sharpened, and the June reaction established that leaving it untouched is not a neutral act. Watch the October-quarter AI guide against the roughly $20.8bn the $56bn full-year figure implies, because that is the step the 2027 number has to build from; group gross margin against the roughly 74% guided, since the semiconductor-versus-software mix behind it keeps moving in one direction; whether the two unnamed customers’ $6bn of orders is given a revenue schedule rather than an order value; and infrastructure software, guided on the call to about $8.9bn against $7.18bn last quarter, or roughly 30% of the guided group total, which is the less-discussed side of the print. The read-across is the broadest of the week – custom silicon, AI networking, hyperscaler capital expenditure, and the Dell and HPE prints either side of it.
🇺🇸 Palo Alto Networks
$PANW reports fiscal fourth-quarter and full-year results Tuesday after the US close, with the webcast at 4:30pm Eastern – the first fiscal year-end with CyberArk consolidated, following the roughly $25bn acquisition that completed on 11 February, less than six months before the year closed. The setup shifted late in the week. The shares closed Friday at $371.59, down 2.94%, but that follows a 12.8% jump on Thursday after CrowdStrike’s results, and the 52-week range is $139.57 to $398.88. The stock is up about 102% since the end of 2025 and about 95% over twelve months, which on consensus puts it near 90 times next fiscal year’s earnings.
The April quarter is the base. Revenue was $3.0bn, up 31%, including $388m from CyberArk and Chronosphere. Next-generation security annual recurring revenue reached $8.1bn, up 60%, of which $1.6bn came from those two acquisitions – organically $6.5bn, up 28%, on organic net new NGS ARR of $370m. Remaining performance obligation was $18.4bn, up 36%, including $1.8bn acquired. The GAAP line moved into loss, largely on acquisition-related costs and charges: an operating loss of $183m against income of $219m a year earlier, and a net loss of $0.22 per share. On a non-GAAP basis operating income was $814m and earnings $0.85, with adjusted free cash flow of $910m and a trailing-twelve-month adjusted free cash flow margin of 38.5%, up 430 basis points. Guidance for this quarter is NGS ARR of $8.90–8.95bn, RPO of $20.9–21.0bn, revenue of $3.345–3.355bn and non-GAAP earnings of $0.96–0.98; the full year is guided to $11.415–11.425bn of revenue, a 28.9–29.2% non-GAAP operating margin and a 37.5% adjusted free cash flow margin. Consensus sits inside the quarterly range at about $3.35bn and $0.98.
Our readThe opening fiscal 2027 NGS ARR guide is the swing factor, not the quarter. The quarter is guided in a $10m revenue band with consensus inside it, so there is little room for the print itself to say much. The guide is a different matter, and it comes with a measurement problem attached: fiscal 2027 is the first year whose growth rate is calculated against a base that already contains CyberArk, so the headline falls from 59–60% to something in the twenties by arithmetic alone. Cantor Fitzgerald put FactSet consensus for that growth rate at 22.1% in mid-August and argued the initial guide has to clear it, which makes the distance between a 21% guide and a 24% guide a far larger signal than the gap suggests; at roughly 90 times forward earnings the stock is not priced for the lower one. Watch organic net new NGS ARR against the $370m posted in April, since that is the line acquisitions cannot flatter; RPO against a $20.9–21.0bn guide; the non-GAAP operating margin against 28.9–29.2% and the 40% adjusted free cash flow margin management has said fiscal 2028 remains on track for; and any evidence that identity is pulling platform deals rather than sitting alongside them. The bar moved on Thursday: CrowdStrike posted record net new ARR of $333m against a $284–286m guide and raised its full-year net new ARR growth outlook, and Palo Alto rose 12.8% on someone else’s numbers. The read-across runs to Zscaler two nights later and to security budgets generally.
🇺🇸 Dell Technologies
$DELL releases fiscal second-quarter results Tuesday ahead of a 3:30pm Central conference call, the same evening as Palo Alto. The shares closed Friday at $456.24, down 3.4% after three sessions that had taken them 9% higher; the 52-week range is $110.22 to $514.00. The twelve-month move is the largest in this week’s calendar – up roughly 240% from a year ago and about 262% since the end of 2025 – and on the company’s own full-year non-GAAP earnings guide of $17.90 the stock trades near 25 times.
Fiscal Q1, the quarter ended 1 May, was a record and a warning in the same set of numbers. Revenue was $43.8bn, up 88%. Infrastructure Solutions Group revenue was $29.0bn, up 181%, with AI-optimised servers at $16.1bn against $1.9bn a year earlier, traditional servers and networking at $8.5bn, up 92%, and storage at $4.3bn, up 8%. Client Solutions Group revenue was $14.6bn, up 17%. GAAP diluted earnings were $5.24 and non-GAAP $4.86; operating income was $3.66bn and cash from operations $4.1bn. The warning is in the gross margin: $7.78bn, or 17.8% of revenue, against 21.1% a year earlier, as AI servers went from 8% of the group to 37% of it. Segment profitability held up better – ISG operating margin was 10.5%, up from 9.7%, and CSG 8.0%, up from 5.2% – but Dell does not disclose an AI-server margin, so that ISG figure is the only direct evidence on the question. The company booked $24.4bn of AI orders in the quarter against $16.1bn recognised, leaving about $51.3bn of backlog. Guidance is $44.0–45.0bn of revenue for this quarter with non-GAAP earnings of $4.80 at the midpoint, and a full year of $165–169bn including roughly $60bn of AI-optimised servers. Consensus sits above both midpoints, near $44.9bn and $4.91.
Our readAI-server margin is the swing factor, and Dell does not report it. Revenue is the easier half of the print: the guide is $44.0–45.0bn, consensus $44.9bn, and $51.3bn of backlog going in makes the top line more a question of scheduling than of demand. What is undecided is what that revenue is worth. Watch group gross margin against 17.8% and the 21.1% of a year ago, and specifically whether the rate steadies now that AI servers are already 37% of revenue rather than continuing to fall as they grow; ISG operating margin against 10.5%, which is the closest thing to a disclosed answer on the question; storage, up 8%, where an improving mix could support ISG profitability as AI servers take a larger share of revenue; and the order and backlog figures, because backlog converted at concession pricing and backlog converted at list look identical in the revenue line and different in the gross margin. The Client Solutions business is worth a glance too – a third of revenue and a quarter of segment profit, growing 17% into a corporate refresh cycle that has nothing to do with the AI story. The read-across runs to HPE the following evening on the same question at a fraction of the scale, and back to Broadcom, which sells into the same buildouts a layer down.
🇺🇸 Snowflake
$SNOW reports fiscal second-quarter results Wednesday after the US close, with the call at 2pm Pacific – the same evening as Broadcom, and the cleanest test in the week of whether AI features are turning into consumption. The stock has the least forgiving setup of the seven. It closed Friday at $328.00, down 0.34%, about 4% below a 52-week high of $341.95 against a low of $118.30, and up roughly 50% since the end of 2025. Most of that came in a single session: $175.26 on the day of the last print became $239.20 the next, a 36% move, and the shares have added a further 37% since. On consensus that is close to 19 times this fiscal year’s revenue.
The April quarter is why. Product revenue was $1,334.3m, up 34% – an acceleration from 30% the previous quarter and 26% a year earlier – on total revenue of $1,391.0m, up 33%. Net revenue retention rose to 126%, remaining performance obligations to $9.21bn, up 38%, and the count of customers with more than $1m of trailing product revenue to 779, up 29%. Non-GAAP product gross margin was 75.1% and non-GAAP operating margin 11.9%, against a GAAP operating margin of minus 23.4%; non-GAAP earnings were $0.39 and the adjusted free cash flow margin 19%. Management raised full-year product revenue guidance to $5,840m, 31% growth, from $5,660m and 27%, and lifted the full-year non-GAAP operating margin guide to 13.5% from 12.5%. This quarter is guided to $1,415–1,420m of product revenue, about 30% growth, at a 12.5% non-GAAP operating margin. The AI disclosure is adoption rather than revenue: more than 13,600 accounts using Snowflake’s AI capabilities, Cortex Code across more than 7,100, Snowflake Intelligence accounts more than doubling sequentially, and a $6bn multi-year agreement with AWS. Consensus for the quarter sits near $1.48bn, but that is a total revenue figure and is not directly comparable with the product revenue the company guides.
Our readProduct revenue is the swing factor, and the quarterly guide is not the test. At $1,415–1,420m it implies about 30% growth, a step down from the 34% just delivered, and simply clearing it would say little. The harder number is the $5,840m full year, raised only one quarter ago: it requires about $4.51bn across the remaining three quarters, roughly 30% growth on the comparable periods, and so has to be reaffirmed or lifted rather than merely met. Watch net revenue retention against 126%, the cleanest read on whether existing accounts are consuming more rather than new logos filling the gap; the non-GAAP operating margin against a 12.5% quarterly and 13.5% full-year guide, since the question underneath it is whether AI-led consumption is growing faster than the cloud-infrastructure cost of serving it; remaining performance obligations against $9.21bn; and the distance between the adoption numbers and the revenue line, because Snowflake does not separate trials, pilots and production workloads in that disclosure, so 13,600 accounts measures adoption rather than monetisation. The valuation cuts against the setup rather than with it: within 4% of a 52-week high and 87% above where it traded on the evening of the last print, the setup asks for a raise rather than a reaffirmation. MongoDB reports the night before on a related question, and Broadcom lands the same evening.
Get weekly earnings previews
Join the waitlist for early access to FinRyse – and receive our weekly market briefings.
🇨🇳 Z.AI
Z.AI’s board meets Monday 31 August to approve interim results for the six months to 30 June – the first half-year figures the company has published since listing in Hong Kong, and the first from a business widely described at its IPO as the market’s first listed pure-play large-language-model developer. The shares closed Friday at HK$1,090, down 6.03%, which is an ordinary day by this stock’s standards: the past ten sessions carried five moves of more than 4% in one direction or the other. It listed on 8 January at HK$116.20 and closed its first day at HK$131.50, so the shares are up roughly eight-fold on that first close; they are also about 63% below an intraday high of HK$2,980, against a low of HK$116.10 – a since-listing range rather than a full year, because January is all the trading history there is. Market capitalisation is about HK$486bn, or roughly $62bn at Friday’s rate.
The only full set of figures the company has published is for 2025, and it frames the question. Revenue was RMB 724.3m, up 131.9% but short of a sell-side number nearer RMB 756m, against a net loss of RMB 4.72bn, 59.5% wider than the year before; the adjusted net loss was RMB 3.18bn. Research and development spending was RMB 3.18bn, up 44.9% and about 4.4 times revenue – growing, but more slowly than the top line. The sharper move was in cost of sales, up 213.3% to RMB 427m against revenue growth of 131.9%, which took gross margin from 56.3% to 41.0% as demand for the GLM models pushed up compute. On Friday’s market capitalisation the shares trade at roughly 575 times that trailing 2025 revenue. Several things have changed underneath since. Bloomberg reported in July that the company has finished building and partly switched on a roughly one-gigawatt data centre stocked entirely with domestically made accelerators, running several clusters of more than ten thousand chips apiece, on which the GLM family is trained; the location, the ownership and financing structure and the share of capacity dedicated to Z.AI have not been disclosed, and 2025 capital expenditure was only RMB 74.7m, with short-term GPU usage booked into research and development rather than capitalised. On the model side, GLM-5.2 arrived in mid-June with a context window of up to a million tokens under an MIT licence, and GLM-5.3-Flash followed on 26 August – a 320-billion-parameter mixture-of-experts model with 18 billion active parameters, native multimodality and the same million-token window and MIT licence, which the company says was served through its Ox Alpha preview entirely on Chinese accelerators. And the first cornerstone lock-up – about 25.7m shares, roughly 12% of the H-share capital but under 6% of the total share count – expired on 8 July, with the shares rising on the day as most cornerstone holders signalled they would stay.
Our readRevenue scale against the loss profile is the swing factor, and there is no robust, like-for-like analyst consensus available for the interim period to measure it by. The published estimates disagree with one another by more than they agree, so this is a print to read against the company’s own history rather than against a number. Watch revenue against the RMB 724.3m the whole of 2025 produced, because a half that annualises well past that is the only commercial evidence available for an eight-fold move in the shares; whether revenue keeps outgrowing research and development, and whether R&D intensity starts to come down from 4.4 times revenue; cost of sales and gross margin against 2025’s 41.0%, since that is where compute lands and it outran revenue by a wide margin last year; and whatever the company says about its infrastructure commitment and the cost of compute – while stopping short of treating the reported campus as an owned, depreciating asset until its accounting structure is known. Domestic-chip efficiency and software utilisation relative to leading Nvidia systems remain the open cost question, and the company has not published a directly comparable performance-per-watt figure. The read-across is wider than the company. This is the closest thing available to a public profit-and-loss account for Chinese AI monetisation, and for the domestic chip stack underneath it. It is also a Hong Kong release landing before the US week begins, on a notice that confirms the board meeting date without stating a publication time.
🇨🇦 Alimentation Couche-Tard
Couche-Tard reports fiscal first-quarter results Tuesday after the Toronto close, with the call the following morning at 8am Eastern – the real-economy counterweight in a week of AI prints, across a network of 17,267 sites as at 26 April, before anything from the pending Żabka transaction. The shares closed at C$83.36, effectively unchanged, and have fallen in six of the last seven sessions from C$87.98 on 19 August; the 52-week range is C$68.30 to C$95.15, leaving them about 12% below the high, up roughly 11% since the end of 2025 and 21% over twelve months. Translated at Friday’s rate, that is about 19 times the $3.10 of adjusted diluted earnings the company reported for fiscal 2026.
The April quarter was strong, and the strength was concentrated. Net earnings were $863.4m, or $0.94 diluted, but that included a pre-tax net recovery of $260.9m from settling long-standing legal matters, mostly US interchange-fee litigation in which the company was a plaintiff; adjusted earnings were $667.0m, up 51.2%, or $0.73 a share, up 58.7%. Wider fuel margins were the main driver, alongside organic convenience growth and acquisitions. The US road-transportation fuel margin was 52.44¢ per gallon, up 9.17¢ year on year, with Europe up US 3.87¢ per litre and Canada up CA 3.23¢ – while same-store fuel volumes fell 2.1% in the US and 4.4% in Europe and rose 2.0% in Canada. Merchandise was steadier: same-store revenue up 3.4% in the US, 1.1% in Europe and down 0.9% in Canada, 2.2% consolidated, at a US merchandise gross margin of 34.4%. Across the year, adjusted diluted earnings were $3.10, up 14.4%; return on capital employed rose from 12.2% to 13.7%; 30.0m shares were repurchased for $1.6bn; and the annual dividend was raised 10.5% to CA 84.00¢. The comparison that matters for Tuesday is the year-ago quarter, which delivered adjusted earnings of $0.78, down 6.0%, on a US fuel margin of 44.00¢ that was itself 4.13¢ lower than the year before. Consensus is around $0.90 on roughly $20.8bn of revenue.
Our readThe US fuel margin against a 44.00¢ base is the swing factor, and it is the least controllable line in the business – set as much by wholesale prices and local competition as by anything management decides. Consensus at about $0.90 asks for roughly 15% earnings growth on the year-ago quarter, and the bridge has to come from some combination of fuel margin, merchandise growth, cost control and a smaller share count. Watch US same-store merchandise revenue against the 3.4% just posted and the 0.4% of the year-ago quarter, since that is the part management actually influences and the part that reads across to the consumer; the fuel margin against both 44.00¢ a year ago and 52.44¢ last quarter, because the second is a high base rather than a run rate; same-store fuel volumes, negative in the US and Europe last quarter, because a wider margin on fewer gallons is a different business from a wider margin on the same gallons; and the European synergy run rate, at about €61m against €120m targeted for this fiscal year. Capital allocation is no longer the quiet part of the story. On 31 July, a little over a year after the Seven & i approach was abandoned, the company agreed to take a controlling stake in Żabka and launched a voluntary tender offer at PLN 32.00 a share, valuing the Polish convenience chain at about $8.6bn – its largest acquisition, backed by hard irrevocable commitments from holders of roughly 57% of the shares, including CVC and Partners Group, and funded by committed debt facilities. Management expects pro forma leverage of about 3.0 times net debt to adjusted EBITDA at closing, a return inside its own framework by year two, and about $250m of annual synergies fully realised by year three, with completion targeted no later than December. The open questions there are tender participation against the 95% needed for a squeeze-out and delisting, regulatory clearance, and whether the buyback slows while the largest deal in the company’s history is paid for. The read-across runs to Lululemon two days later on how the North American consumer is spending, with the caveat that a fuel-and-convenience network and an aspirational apparel brand are asking rather different questions of the same wallet.
🇺🇸 Lululemon
$LULU reports second-quarter results Thursday after the US close, five days before Heidi O’Neill takes over as chief executive on 8 September, ending a stretch run by interim co-chief executives Meghan Frank and André Maestrini. It is the most asymmetric print of the week. The shares closed Friday at $120.81, up 5.05%, about 16% above a 52-week low of $104.44 and roughly 47% below a high of $225.98; they are down about 42% since the end of 2025 and by a similar margin over twelve months. Against the company’s own full-year guide of $10.95–11.15 that is about eleven times earnings, the lowest of the seven companies covered here.
The first quarter is the base, and the split inside it is the story. Revenue rose 4% to $2.5bn, or 2% in constant currency, but Americas revenue fell 3% and Americas comparable sales fell 5%, against international revenue up 22% and international comparable sales up 13%. Gross margin fell 410 basis points to 54.2% and operating margin 730 basis points to 11.2%; diluted earnings were $1.69 against $2.60. Inventories rose 2% in dollars while falling 4% in units. The company bought back 2.2m shares for $358.3m and ended the quarter with 816 stores and $1.5bn of cash. Guidance came down with the print: $2.450–2.475bn of revenue for this quarter, down 2% to 3%, with earnings of $1.76–1.81, and a full year cut from $11.35–11.50bn to $11.000–11.150bn, a decline of 1% to flat, with earnings cut from $12.10–12.30 to $10.95–11.15. Management attributed the cut to traffic hit by “spikes of negative commentary in the media and on social channels” and to product launches that did not land, and guided North America down low double digits for the quarter against China Mainland up mid-to-high teens. Full-year gross margin is guided down about 90 basis points and operating margin about 380, with tariffs costing roughly 30 basis points of gross margin across the year after mitigation – but about 150 basis points of gross pressure in this quarter alone, against roughly 100 basis points of planned offsets, or about 50 basis points net on management’s assumptions. Consensus sits mid-guide at about $2.458bn and $1.79.
Our readWhether the Americas decline is stabilising or still widening is the swing factor, and the direction is already guided. North America was guided down low double digits, so a weak number is not new information; what is undecided is the second derivative, because a low-double-digit revenue decline is a marked step down from the 3% fall in Americas revenue in the first quarter, and anything at the better end of it would be the first evidence the reset is finding a floor. Watch North American revenue against the low-double-digit decline guided and Americas comparable sales against the minus 5% just posted, keeping the two apart – they are different metrics over different geographies, and the company did not guide a regional comparable-sales figure for the quarter; gross margin against 54.2%, bearing in mind that this quarter carries about 150 basis points of gross tariff pressure the company has flagged as the year’s peak; markdowns, which management called the high-water mark for the year in this quarter; and inventory, where 2% more dollars on 4% fewer units means more value per unit, though that on its own does not separate price, mix, input cost and currency. The full year matters more than the quarter. At $11.000–11.150bn and $10.95–11.15 the bar is already low. Heidi O’Neill starts five days after the print and may revisit the medium-term priorities once in office, but any reset on the night would be the current team’s to make. That is the asymmetry: expectations are modest, the stock has given back two-fifths of its value this year and trades near eleven times the bottom of its own guide, and the risk of another cut has not gone away. The read-across runs to aspirational athleisure and discretionary apparel generally, and back to Couche-Tard two days earlier on a different corner of the same consumer.
Weekly Calendar
Some notable names reporting this week. “Post-close (exp.)” marks a confirmed reporting date with no stated publication time, where the timing follows the Hong Kong convention rather than the company’s own notice.
| Company | Country | Sector | Timing |
|---|---|---|---|
| Monday 31-Aug | |||
| Z.AI | 🇨🇳 | Large language models | Post-close (exp.) |
| Tuesday 1-Sep | |||
| Palo Alto Networks | 🇺🇸 | Network & platform security | AMC |
| Dell Technologies | 🇺🇸 | AI servers, storage & PCs | AMC |
| Alimentation Couche-Tard | 🇨🇦 | Convenience stores & fuel | AMC |
| Credo Technology | 🇺🇸 | High-speed connectivity | AMC |
| MongoDB | 🇺🇸 | Developer database platform | AMC |
| Partners Group | 🇨🇭 | Private markets asset management | BMO · 07:00 CET |
| Wednesday 2-Sep | |||
| Broadcom | 🇺🇸 | Custom AI silicon & infrastructure software | AMC |
| Snowflake | 🇺🇸 | Cloud data platform | AMC |
| Hewlett Packard Enterprise | 🇺🇸 | Servers & networking | AMC |
| NetApp | 🇺🇸 | Enterprise storage | AMC |
| Thursday 3-Sep | |||
| Ciena | 🇺🇸 | Optical networking | BMO |
| Zscaler | 🇺🇸 | Cloud security | AMC |
| Samsara | 🇺🇸 | Connected operations software | AMC |
| Guidewire Software | 🇺🇸 | Insurance software | AMC |
| Lululemon Athletica | 🇺🇸 | Athletic apparel | AMC |