ANET -Jayshree Ullal: Queen of the wired world

Interesting article on the CEO.

Jayshree Ullal: Queen of the wired…

This link might work better. Looks like an article from 2013 (but surely still includes some interesting points for anyone that might have considered investing in Arista Networks).…


I read this last night. The Ullal family is definitely quite successful/interesting in business. I had no idea prior to reading this article that Jayshree’s husband Vijay had been the President of Fairchild Semiconductor (up until about 2014, I think).

Reading the article also prompted a bit of further research. Their older daughter Adeeti has a PhD from MIT/Harvard Medical School and works for Apple now as an engineering manager presumably in connected fitness and has done some work to improve detection of lung cancer. Their younger daughter is still in vet school, I think.


This caught my eye from my Arista Networks Google Alert. I guess Jayshree simply needed about $50M…maybe to buy Divali or Christmas presents for her kids or invest in some sort of other venture (or to buy a new house or boat)? Likely not of any particular note for shareholders, and the market doesn’t seem to have had any issue with it, as the price this morning is up again (now over $230/share, having started to leave NVDA in the dust a bit on a share price basis since crossing paths at $212…still far behind NVDA in market cap).

The headline number says $16.4 million, but the article lists three separate sales on 11/8, 11/10, and 11/13, which totaled $49.8 million from my calculation.…

New Gulf Stream is 65 million.

1 Like

She can buy several Gulf Streams. She has been systematically selling ever since the company went public. She has lost an awful lot of money because of these automatic sales. Her selling has never had any correlation to company performance. They are automatic sales, and she will keep selling into perpetuity.


Is the programmatic selling simply built into her compensation package, Tinker?

I knew better than to be at all alarmed by it, but it did catch my eye as Arista has moved into my being probably my most confident position looking forward (partially due to NVDA now being at an almost $130B market cap).

I do not know the formal process. But I have looked back historically and she has sold more than $100 million dollars in stock options since the IPO with no correlation at all as to the company business.

How much she netted from all this, I do not know. She has to pay the strike price and I do not know what the strike price is on these options, so she is not netting the gross amount, plus she has to pay taxes. It is not unusual at all for executives to systematically sell as that is the bulk of their compensation.

But it appears to be automatic selling so as to avoid any insider trading allegations, and also any tax consequences from holding stock options as part of her compensation and not having the money to py the taxes on them at the end of the year.

It caused me concern when I first bought ANET as well until I looked back at her very large insider sales.



$45 million. Chicken feed for whoever bought this painting…

If she keeps selling all her shares as ANET grows from a several billion dollar company to a company worth several scores of billions, she’ll have to settle for being filthy rich instead of filthy stinkin rich. Poor thing.

I say let her do whatever and pay her whatever she wants. She’s done amazing things with Arista and they just continue kicking Cisco’s tail.

added to ANET yesterday

1 Like


Are you following any kind of strategy when adding to ANET or just adding when it feels right? Maybe added when it dipped a little yesterday?



Are you following any kind of strategy when adding to ANET or just adding when it feels right? Maybe added when it dipped a little yesterday?

Just when the spirit moves me. Ha…no, I’m no market timer and I know nothing of trading or technicals. I just like the value prop right now with Arista.

As for yesterday, maybe because it was down, I decided to revisit / look admiringly on the company results, and I said to myself, “The way Arista is going I could easily see them having $8 or more in EPS in 2018…the market is expecting $6.12? I better add some more shares.”

That’s all there was to it. Certainly not a buying “strategy” or anything scientific. But I think I know a good company when I see one. And I’m willing for this to be a very large position while the company is growing like it is, and especially while the stock valuation is where it is (PE = 47 currently).



It is amazing how far behind analysts are, particularly when it is their full time job to cover these things. I have owned stocks that I’ve known very material things about for 6 months or longer, and the market never paid heed until the right analyst or magazine actually came out with the news. Lets just say the market is mostly efficient, but is still a human institution.

Here we have Arista now eating into Cisco in the enterprise markets, its disruptive routing product eating into Cisco in routers, and the analyst uses the term: "In enterprise accounts, in particular, writes Henderson, Arista is showing “exceptional” growth. "Many of these customers are looking to the scaled out cloud model and want to emulate it,” observers Henderson.

“Post the Cisco call we think its pretty clear Arista is starting to bite into both Switching and Routing market shares in Enterprise and Service Providers.” Henderson is referring to Cisco’s conference call yesterday to discuss fiscal Q1 results and the Q2 outlook. Cisco’s sales of routers declined by 6%, and sales of switches declined by 4%, he notes.”

Those declines are “stark,” writes Henderson.

Both Juniper and Cisco reporting declining sales while Arista reports accelerating sales with their highest growth rates as a public company. Quite stark the difference.

All this during the 100gb inflection point. However at some point even Arista’s growth may slow, but that happens to every company.

But this is why I moved heavily into ANET. 100gb inflection point, router disruption, and large untapped opportunity in enterprise and international and Arista is doing all this with only 10% or less of revenues going to marketing. It is clearly a product and architecture that sells itself. More impressive as it steals marketshare from large and very established incumbents, all in what was considered a mature and staid market.

The thing with Arista, all this was telegraphed 18 months ago as what Arista does now is usually not seen for 6, 12, 18 months. It was not more than 2 or 3 earnings calls ago that Arista announced that their partnership with HP to sell into the enterprise had finally hit the stage where enough training and coordination had been done that HP would be able to start really selling the product within the next few quarters.

Apparently the analysts was not paying attention.

Anyways, thought it interesting to share from a figure with “authority” those things that we already figured out for ourselves.

Reminds me of the sell call on Nvidia despite the tornado in data center AI and really no competition.