Devendra

[Devendra @_Devendra_Govil]

@pranaykotas @iid_sp In that sense, it's definitely a card with an expiry date whose expiry countdown has started, but useful enough to get atleast some leverage in negotiations for the short to medium term.

[Devendra @_Devendra_Govil]

@pranaykotas @iid_sp They are not meant to be permanent, just a tactic to buy time and have some cards to negotiate with. The estimate of a parallel supply chain is up in the air, but it's not as trivial as some people have made it sound.

[Devendra @_Devendra_Govil]

@TheMinuend @cvkrishnan These are not mulberry barges by a long stretch. No one even claims that, just that they are "similar" in function to mulberry barges.

These are the tip of an amphibious attack. They will only be used when a beach head is secured, and after a long air campaign.

[Devendra @_Devendra_Govil]

@akapoor_av8r Exceptional thread! Thank you so much for providing these insights.

[Devendra @_Devendra_Govil]

@Sheikh__Zohaib @pstAsiatech @FT @tphuang Any sources?

[Devendra @_Devendra_Govil]

@teortaxesTex I think they need a certain level of diversity in industries represented. It is a meeting with private entrepreneurs, not just tech. Tech (of all kinds) is already well represented.

[Devendra @_Devendra_Govil]

@VyvyenYue Don't worry. Trump seems to be set to un-ban everything on Jan 20.

[Devendra @_Devendra_Govil]

@AvisharDutta @Mohansinha @DivaJain2 You keep grinding. Keep the struggle on. You develop C929, start study for C939. Develop CJ2000, start developing the industrial chain, while scaling production of C919 to 150 aircrafts/year by 2027. You spend sweat and tears and blood to learn and iterate, rinse and repeat.

[Devendra @_Devendra_Govil]

@AvisharDutta @Mohansinha @DivaJain2 Then further to small turbojets like (C909), now to narrowbody airliners like C919 that are good enough. No doubt C919 is not competitive in head to head match, but you persevere, you develop the CJ1000A to have within 5% of the fuel efficiency. You don't stop.

[Devendra @_Devendra_Govil]

@Sheikh__Zohaib @tphuang @RnaudBertrand Just curious: How is weichai dominating? Any sources/reports

[Devendra @_Devendra_Govil]

@kyleichan @IndustrlPolicy The past few years inflection point is when the world starts to observe it. I don't think there has been any inflection point, the overall move to quality and innovation is inline with per capita GDP, stage in industrialization process, and research capacity.

[Devendra @_Devendra_Govil]

@damienics @gonglei89 thinnest solar wafers in the world. It's hard from a process perspective to make them thinner (think lower node lithography, or ICs). But thinner solar wafers is straight cost reduction.

[Devendra @_Devendra_Govil]

@damienics @gonglei89 Lithography has been available since >5decades, doesn't mean it is old tech. Even EUV lithography has been known in lab since late 1990s. Same for IC, just because ICs have been available since 1960s doesn't mean that ICs is not advanced. Solar has made incredible gains.

[Devendra @_Devendra_Govil]

@negativeonehero Are there any teardowns and component analysis available for P70?

[Devendra @_Devendra_Govil]

@kyleichan Both are talking about very different things. The article talks about semiconductor assembly and packaging, which is an entirely different beast than electronics assembly. Semiconductor packaging is even done by TSMC so can be highly technologically intensive!

[Devendra @_Devendra_Govil]

@dunne_insights More imp. are infact their M3P, Shenxing and (the pack level innovation) Qilin Batteries. M3P is LMxFP, with LFP augmented with Manganese and a few (as yet undisclosed) elements. CATL claim -> competitive with NMC in vol energy density while retaining many benefits of LFP.

[Devendra @_Devendra_Govil]

@pstAsiatech @NikkeiAsia This evolutionary word is extremely misused to hype up research. There's a class of optimization algorithms called evolutionary algorithms (one is genetic algorithm) which sound fancy but are nothing like that.

[Devendra @_Devendra_Govil]

@wandb "Model Evaluation, Model Selection, and Algorithm Selection in Machine Learning" by Sebastian Raschka is one of my go to papers to review all details of model evaluation etc. Available at https://t.co/iN6wsCkVOK

[Devendra @_Devendra_Govil]

@ashajadeja Can I know the context here? I seemed to have missed something.