2026 YTD Monthly Closing
Jan: -5.5%
Feb: -17.5%
Mar: -20%
April: -9.4%
May: +19.5%
June: +22.4%
July: +15.9%
Aug 31st: +26.4% YTD, (down from a YTD peak of 30% in August but up from the YTD low on March 30th of down 24%). One point to note is that the I have been withdrawing from the portfolio to a now significant degree in the last few months leaving the gross total value gains and the adjustment for withdraws at about an equal contribution to the normalised YTD increase.
August saw a strong rebound through the month whilst running up against significant macro event driven downdrafts. Whilst AI infra/connectivity, (supported by very strong Q2 results announcements) made a comeback together with SaaS, (with favourable consumption based business models, hardware attach and proven agentic AI beneficiaries), neoclouds were left behind in this rally. To really go beyond the ATH and make substantial further progress my portfolio is going to have to rely on both engines instead of only one at a time!
I was happy enough with the SoFi, Pagaya and Robinhood Q2 announcements and very impressed with Reddit, Silicon Motion, Bloom Energy, Palantir, Astera Labs, MercadoLibre, Figure, Axon, Crowdstrike, Shopify and Everpure and whilst not holding direct exposure to the hyperscalers, Microsoft was a stand out to me. Tempus AI, Toast, SentinelOne and IREN were a meh for me at this point and AppLovin and Rubrik looked problematic to assess.
Thematically, I’m principally invested in:
eCommerce (15%) - Shopify, MercadoLibre, Global e-Online & SEA
AI & Cloud Infrastructure (30%) - Cloudflare, Pure Storage, Nvidia, Nebius, Astera Labs, Micron, Tempus Ai, IREN, HIVE, CoreWeave, Credo, ORCL, Lumentum & EOSE
Software (SaaS/DevOp/Data analytics) (25%) - Palantir, Datadog, Snowflake, GitLab, Monday, Axon & Samsara
Cybersecurity (10%) - Crowdstrike, ZScaler & Rubrik
Fintech/Payments/Crypto (10%) - SOFI, Toast, Robinhood, Upstart, Figure, Pagaya & Bitmine Immersion Technologies
AdTech (10%) - The Trade Desk, Applovin and Reddit (ad community)
Recent Activity -
In August I trimmed Astera and Credo at their peak but have since added back to both. I trimmed Shopify on its run up and trimmed Everpure going into earnings (but since then bought some back - I will post separately on this as I see a potential Nvidia/Micron moment upon us there) and considering trimming Palantir.
I felt it was high time I got ruthless about some of my low conviction lower growth holdings that were stubborn in their turnaround or value realisation, including: SentinelOne (who lost a growth advantage to peers and now under-grow whilst still the smaller/younger player) and BILL (which never realised its takeover speculation nor turnaround potential).
I entered Lumentum as an addition and I’m considering entering: ONTO, Coherent & Rocket Lab, Billion To One and some energy storage players like Bloom or Fluence or NRGV as well as Silicon Motion and Dell (which I should have pulled the trigger on). Drone makers and their eyes/ears plug in support with recurring revenues could also be an area of interest.
Holdings in Monday, GitLab, Toast and TTD sit closest to the exit door.
Portfolio holdings -
25+ positions with a long tail of 1-2% positions (made up of high conviction, scaling down and scaling up plays):
MU - 10%
SHOP - 7.5%
DDOG - 6%
NBIS - 6%
NET - 6%
PLTR - 6%
P - 5.5%
SNOW - 5%
CRWD - 4.5%
SNOW - 4%
ALAB - 4%
NVDA - 4.0%
RBRK - 3.5%
IREN - 3.5%
MELI - 3.0%
RBRK - 3.0%
IOT - 2.0%
ZS - 2%
SOFI - 2.0%
HOOD - 1.5%
CRWV - 1.5%
APP - 1%
GLBE - 1%
FIGR - 1%
LITE - 1%
CRDO - 1%
MNDY - 1%
DLO - 1%
GTLB - 1%
AXON - 1%
UPST - 1%
ORCL - 1%
RDDT - 1%
TEM - 1%
SE - 1%
PGY - 1%
HIVE - 0.5%
TOST - 0.5%
BMNR - 0.5%
TTD - 0.5%
EOSE - 0.5%
Bright spots in the portfolio include: Nvidia as well as Cloudflare, Crowdstrike, Shopify, Palantir, Snowflake, Lumentum, Rubrik, Gitlab, Samsara and Datadog which are closest (within 20%) to their 52W and AT highs.
Watch list includes…
Fluence, Bloom, Sigma, Coherent, Celestica, Rocket Lab, Silicon Motion, Onto, Sezzle, ROOT, FOUR, Arista, Fortinet, Palo Alto, Raspberry PI and Grab.
Bigger Picture -
As sectors, Cyber Security, AI and Cloud infra/DevOps have been relatively strong - as well as profitable SaaS/software providers that operate a platform play with consumption based revenue models and demonstrably bullet proof vs AI disruption. Payments & Fintech which had been strong with rate reductions back on the cards, expectation of spending resilience and a boom in crypto/alternative/private investing, however that strength has weakened in recent months with economic uncertainties (that are also impacting eCommerce players), whilst AdTech is showing signs of weakness in pricing and demand as well as facing a fight back from the walled garden operators going beyond their native home markets.
It feels as though every part of the data space is doing well and reaccelerating; (semiconductor, memory, storage, servers, data center infra, databases & data cloud) and remains AI resilient. Clearly the most extreme gains have been seen in anything that could constitute an AI value chain supply side bottleneck
I see the outcomes of 4 challenges are the critical determinants of market success right now - certainly for my portfolio holdings:
i) the will it won’t it question of “AI eating software”
ii) the alternative scenarios of higher for longer vs an AI bubble
iii) the formulation and reaction to Macro (tariffs & trade), fiscal (tax & spend) & monetary (Fed rates) policies
iv) the imminent behemoth IPOs (SpaceX, OpenAI & Anthropic) and to what degree that will create investor euphoria across the board or suck money out of the market in order to fund IPO entry positions
Specifically the almost binary outcome of OpenAI and its potential impact on NeoCloud Capex, Oracle RPO commitments (amongst others) and leadership in AI feels a critical risk/reward situation.
One additional competitive risk I am watching carefully is the transformation of X from a social media platform to a fully fledged all-in-one super app with eCommerce and Fintech which could impact a number of my holdings from Shopify to SoFi within US and MercadoLibre and SEA internationally.
I still believe that seeking out growth companies with defensive qualities, (cloud infrastructure, cybersecurity, energy generation, storage and supply even and consumption based rather than seat based software models), might do better in these volatile and uncertain times as well as ex US eCommerce and trading plays that benefit from US currency weakness but more importantly are removed from US import/export movements on an intra regional / local to local basis (e.g. MercadoLibre, SEA and Grab).
Ant