operry retuiteado
operry
1K posts


@JamesMac_Fit No iPhone or iPads until 16… they’re allowed to watch TV after they clean up their mess. Max of 1hr per day, and only 3 days per week.
English

Dads... I need your advice.
As a young dad myself, I’m genuinely worried about my son growing up addicted to screens, phones, games, etc.
It feels like EVERYTHING is designed to grab their attention these days.
So I want to ask:
- How do you actually manage screen time in your house?
- What’s worked for you?
- What’s failed?
Would appreciate real answers from people who’ve been there 🤝
English

Ceres Power is an alternative to $FCEL + $BE:
We all know about the power bottleneck for data centers by now?
Solid oxide fuel cells (SOFCs) bypass current grid limitations by offering rapid time-to-power.
Ceres, $BE & $FCEL are all in SOFCs:
$BE wins in the sector due to their commercials (e.g. $NBIS, $ORCL) + is largely priced-in now imo.
But I think Ceres comes in second place when looking from a tech lens.
I’m not going to trash talk $FCEL cos I actually do have a position w/ a ~70:30 split weighted towards Ceres.
And just quickly: I ultimately seeing the whole Fuel Cell sector winning together since the TAM is so huge.
Note: Being transparent, I'd be very shocked to see any sort of parabolic run like $SIVE type names.
---
With Ceres, they abandoned attempts to manufacture + pivoted to a pure IP-licensing model:
They’ve developed next-gen + patented solid oxide tech IP.
This enables Ceres to license IP to manufacturing partners for mass production. They also license system IP where stacks are integrated into power systems, which are sold to end markets.
They then earn revenue (royalties) through up-front licence fees for access to the IP + are based on kW of product sold into end markets.
This model delivers high margins (~70%) + recurring revenue + minimal capex for Ceres itself.
FY25 revenue was £32.6M (first-ever royalties of £110k from Doosan)
And FY26 has £45M already contracted.
So really, the investment case hinges on whether Ceres can convert its tech leadership + partner momentum into scalable royalties.
And yes, I do believe they’re leading r.e. the tech itself vs. other SOFC companies:
They launched Ceres Endura in April which claims:
-> 5-year stack life
-> >65% electrical efficiency on nat gas
-> >90% total efficiency in CHP applications
-> "up to one-third lower system manufacturing cost at scale than alternative high-tempSOFC technologies"
In plain English - Ceres runs at lower temps (450-630C) vs. $BE (800C) vs. $FCEL (~650C).
Every other technical advantage flows from temperature, because temperature drives degradation kinetics + BOM.
I’m not gonna pretend to fully understand the tech, but that’s the TLDR to the best of my knowledge.
---
Field data from Doosan deployments over 2026-28 will be the decisive validator, but the design directly targets Bloom’s cost structure while maintaining the licensing model’s operating leverage.
Ceres’ value creation ultimately rests on its manufacturing partners, who have collectively sunk an estimated ~$850M in factories tooled to its IP.
Key relationships include:
-> Doosan Fuel Cell (Korea): 50MW factory went live in July 2025 + generated Ceres’ first royalties in FY25. 2026 ship volumes (targeting tens of MW) are the most important near-term catalyst for Ceres’ re-rating.
-> Delta Electronics (Taiwan): Signed licence in 2024 (£43M over three years). Delta has acquired land for a factory (~£170m investment) + formed a strategic partnership w/ Centrica (UK) to deploy off-grid SOFC systems for UK + European data centres. First commercial product is targeted for end of 2026.
-> Weichai Power (China): Nov 2025 manufacturing licence agreement gives Ceres access to the Chinese market for AI data centres. Weichai are a ~20% shareholder + ~$30B rev giant.
All this to say, things are still early for Ceres:
$BE ships gigawatts today on older, hotter, less efficient chemistry. $FCEL ships about 50 MW per year of structurally inferior technology.
Ceres has shipped £110K of royalty rev (to Doosan) across all partners in its 25-yr history.
So Bloom has the best commercial biz by far.
For quick napkin maths:
-> A 2x re-rating on 15x EV/sales requires ~£200M revenue.
-> At 2-4% effective royalty on stack value (~30% of system) that's ~$25B cumulative sales.
I don’t see that happening at current partner capacity, so would potentially need new licensees signing, not just existing partners scaling.
Which is definitely possible according to mgmt commentary:
Ceres’ addressable SOFC market is estimated by management at ~22 GW by 2030 (roughly 50% data centres, 50% industrial/commercial).
The royalty model means Ceres captures upside with limited capital outlay.
A single partner scaling to hundreds of MW can drive tens of millions in annual royalties at plausible 2-4% effective rates on stack value.
Multiple partners ramping + new MLAs could push revenue into low hundreds of millions by the early 2030s.
---
But there are ofc risks e.g. dilution:
Share count grew from ~80M (2015) to ~195M (2026). Future dilution risk currently low imo (cash runway to 2029) but not zero if M&A / capacity buildout pivots strategy.
Another is the licensing model itself can be a drawback since they're at the whim of their partners' operations.
So really, you'd wanna track their partners closely to see what deals they're signing in the future.
---
Disclosure 1: as mentioned, I currently have a position. Mainly out of support for UK tech.
Disclosure 2: if I need the cash for something else, I may sell due to already being invested in $BE, so it could easily turn into a short-term swing trade.

English

While I have a position in $NBIS, and I think it's the highest quality in this space, I have been puzzled why $CRWV hasn't been getting more love.
Is it that the market pref dilution over debt by that much?
Or is it looking very far ahead to where the neocloud that Nebius This building is a key differentiator, which I think will absolutely matter in 4 or 5 years?
Or is there something else I'm missing?
Either way, I might be shifting from Nebius to Coreweave soon.

English
operry retuiteado

@garyblack00 All the people saying it’s overpriced are making me more bullish by the day
English

@gregmushen Many of the datacenters recycle their water in a closed loop, fyi
English

@DarkPoolTA @saxena_puru Agree. It’s the most deflationary force we’ve ever constructed.
English

@saxena_puru 6-8 months* and it isn't an "AI Bubble" it is an intersection between government debt levels and inflation. AI is inherently disinflationary.
English

@saxena_puru 1-2 years? You’re retarded. This is a fundamentally altering technology that we’re just starting to apply to our lives. Less than 10% of the world uses an LLM daily.
English

When does circular financing stop? Similar to a Ponzi Scheme, when cash outflows become greater than inflows. In this last quarter, $NVDA cash grew by only ~$600, and now ~95% of its operating cash flow is absorbed by circular financing vs ~57% one year ago

JustDario@DarioCpx
Going to sleep now, enjoy the upcoming granted “beat and raise” usual $NVDA earnings charade. When I wake up I will take a close look at the earnings and if I find anything interesting nobody already posted I am going to share it as I have been doing for the past 4 years. 👋
English
operry retuiteado
operry retuiteado
operry retuiteado

@aleabitoreddit I think it’s more about fewer people turning to Reddit for advice and increasingly relying on AI
English





















