Sydney☄️🇰🇪

CEO @hussiniNairobi |intelligent capital market infrastructure
2038🐦‍🔥.

[Sydney☄️🇰🇪 @Kuradul]

DRC has no capital market yet.
Although we expect a soft launch of one in 2027.

[Gichuki Kahome @kahome_steve]

1. Remember also Equity as Dr James Mwangi mentioned intend to have other subsidiaries starting with DRC start paying dividends. This will be huge for Equity as it is expected that the DRC subsidiary(which is growing faster than the kenyan unit) will surpass the Kenyan unit in profitability by 2028

2. When valuing a bank. The justified PB adjusts PB using ROE...Equity Group not only has a superior ROE to KCB, but its ROE is also superior to its Cost of Equity....This in the valuation methodology earns it a higher premium to book than KCB.

[Sydney☄️🇰🇪 @Kuradul]

Ignoring NCBA is a massive mistake.
Accumulate.
Management has a specific platform thesis that will pay off.

[Sebastian Raschka @rasbt]

A lot of hype around OpenAI's Astra model here on my timeline today. Apparently, this goes back to a new article from The Information, which said Astra is a "recurrent depth or looped transformer".

It's always interesting to read about new or different approaches (including rumors about what the closed labs may be up to), but let's debunk this a bit.

About 2 months ago, I shared the architecture details of Nanbeige, for example, where "Nanbeige4.2-3B is pretrained from scratch on 28T tokens with a Looped Transformer that reuses the layer stack to increase capacity without adding parameters."

Yes, that's it. The looped transformer idea is just reusing layers in the transformer block.

In the case of Nanbeige, the main idea is to reuse the same 22-layer stack (=transformer block) twice instead of once. So, effectively it extends the 22-layer architecture to 44 layers, but without duplicating the weights.

In simple terms, this roughly doubles the size of the model (if we ignore the embedding and output layers for a second). But instead of requiring 2x the storage and RAM to host this model, it stays at the same size since we reuse the components. However, it's almost 2x as expensive in terms of compute, because we run the embedded text through almost 2x as many layers.

Why? In the Nanbeige 4.2 technical report, the researchers found that two passes gave the best trade-off and retained about 75% of the token efficiency of a standard architecture. (More passes gave barely any gains but made the training much slower and much more expensive.)

While, as far as I know, Nanbeige 4.2 is the first notable open-weight model that adopted this approach, the idea goes back to the NeurIPS paper "Mixture-of-recursions: Learning dynamic recursive depths for adaptive token-level computation". Actually, this paper proposes a mechanism that is a bit more sophisticated by adding a learned router that determines whether each token receives one, two, or more passes. So, easy tokens can exit early while harder tokens receive additional computation.

In sum, Astra may be a really good model, but this shouldn't be about this "looped transformer aspect," which is just a tiny architectural tweak.

Also, the statement "the new technique works in a way that obscures some or all of the AI's reasoning, otherwise known as 'chain-of-thought'" is not necessarily true with respect to the looped transformer method. It's possible that The Information journalist refers to some other technique or misunderstood the looped transformer method.

Reusing layers does not by itself suppress visible chain of thought. It adds computation in hidden states before the next token is emitted, just as ordinary transformer layers do.

But based on the information we have, the only plausible interpretation here is that if a model uses more of these recurrent passes, it may need to generate fewer intermediate reasoning tokens. So then more of its computation happens in latent activations that cannot be read as text. But we would get the same effect if we were scaling up the model size, like GPT 5.6 Luna -> GPT 5.6 Sol.

[Sydney☄️🇰🇪 @Kuradul]

The Takeaway:
An acquirer is buying the CMA license, tech rails, and KES 530M+ CIS AUM, not cash flow. Expect a steep haircut on receivables and a 40%+ earn-out tied to AUM retention

[Sydney☄️🇰🇪 @Kuradul]

Balance Sheet Reality Check:
Receivables Trap: KES 45.3M (50.5% of assets; DSO ~271 days).
Thin Treasury: Cash at KES 5.3M (~$41K); short-term investments dropped from KES 51.2M to KES 0.
CAC Drag: Marketing consumed KES 17.0M (56% of revenue) for KES 1.36M PAT.

[Sydney☄️🇰🇪 @Kuradul]

@felix_odhiambo_ 1 kg of meat is 1000 kes now.

[Sydney☄️🇰🇪 @Kuradul]

@johnam85 The counter is well positioned, look at its Fx platform and the remittance partnerships they have contributing significant flows regionally.

[Sydney☄️🇰🇪 @Kuradul]

I&M is important when looking at its integrations with remittance firms anchored in those flows across the region.
One counter i love its busienss model.

[Sydney☄️🇰🇪 @Kuradul]

I believe in future this drc unit will overtake the kenyan unit by size and earnings in the next 25 years.

[Sydney☄️🇰🇪 @Kuradul]

Its the fastest growing subsidiary by profit earnings all its subsidiaries.
Biggest bank in DRC , just to visualise this bank is also the oldest in the country and 90% of the population is still unbanked with kinshasa having 15 million+ people alone.

[Business Daily @BD_Africa]

DRC biggest gainer as Equity Group hires 410 more workers
https://t.co/NIfUw4UCbE

[myTradeSignals @mytradesignals]

Total Energies $TOTL HY-2026

PAT = +21.3% YoY

[Sydney☄️🇰🇪 @Kuradul]

He mentioned a hedging capacity on live TV, wr need this product clearly defined and regulated as a hedge fund.

[Standard Investment Bank @investwithSIB]

📈Long-Short Trading Model that works for Mansa-X

Why does Mansa-X Special Fund deliver consistent returns?

In this episode of @ntvkenya's Business Redefined hosted by @AmbokoJH, @_NahashonMungai, SIB Executive Director, Global Markets and Portfolio Manager of Mansa-X Special Fund, explains the investment strategy behind Mansa-X’s consistent performance across different market conditions.

He highlights how the fund’s long-short trading model provides a built-in hedge, allowing the portfolio to manage market exposure and pursue more consistent returns, regardless of market direction.

Catch this full conversation in the link below where we dive into the Long-Short trading model and why it works for the Mansa-X Special Fund.
📈https://t.co/fNxzOhSrNd

#MansaX #BusinessRedefined #investwithSIB #NTVKenya

[Sydney☄️🇰🇪 @Kuradul]

As i mentioned , a diversified portfolio doesn't mean owning multiple stocks.
Practically you want to own different asset classes.
Invest with conviction do your own research and size up.

[Trader Theory @tradertheory]

Billionaire investor Peter Lynch on how diversification is useless.

[Trader Theory @tradertheory]

Billionaire investor Peter Lynch on how diversification is useless.

[Sydney☄️🇰🇪 @Kuradul]

@FabrizioRomano Szobo?

[Sydney☄️🇰🇪 @Kuradul]

Wild stuff, just tried it .

[Fastlane @UseFastlane]

Meet Duolingo for Marketing.

Just enter your website URL, and Fastlane will create thousands of accounts and videos promoting your product in seconds.

This is honestly crazy.

[Sydney☄️🇰🇪 @Kuradul]

I like the idea , although its a black box , give us more control over the data .
An affordable price and data feed api would be nice🫠

[Rock Investment Bank @Rib_investment]

Financial data shouldn’t be hard to find.

Rock Data Plus brings NSE, market and macroeconomic data together in one place for students, researchers and professionals.

Find it. Download it. Analyse it.

🔗 https://t.co/wMYJTEhuVZ

#RockDataPlus #RockInvestmentBank #MarketData

[AXYS INVESTMENT BANK @AIB_AXYSAfrica]

Government Bonds Available on Digitrader

Investors can currently access the following Government Bonds through Digitrader:

• IFB1/2021/18
• IFB1/2021/21

[The Kobeissi Letter @KobeissiLetter]

BREAKING: Bitcoin rises above $80,000 for the first time since May 15th.

Bitcoin is now up +29% in 7 days.

[Sydney☄️🇰🇪 @Kuradul]

Dreams coming true , thank you. @WallStAfrica @ericasuma

[Sydney☄️🇰🇪 @Kuradul]

Less REITs more stock ETFs on the bourse please.