@CodeAndCrease
Account based inIndia
About this account
- Account based in
- India
- Connected via
- India Android App
Account-level information from X, not a live location or the device used for a specific post.
Pinned Tweet
Two things I live between:
The crease — where one ball can change everything 🏏
The code — where one bug can change everything 💻
Welcome to @CodeAndCrease . India's messiest overlap starts now 🇮🇳
This guy didn't even tag @DKShivakumar (who made it possible) in fear of not offending someone else 🤡
#Karnataka #Tulu
503/9 declared. Enforced the follow-on. Still couldn't win.
Sri Lanka's Dinusha (103 & 133*) batted 426 balls across two knocks to save a Test India dominated for four days. This is now Gambhir's Test record: 7W-10L-2D in 19 games. WTC final hopes? Gone. Again.
Dominance without a scoreboard result isn't progress, it's a pattern. #INDvSL #TeamIndia
I'd never UNDERSTAND someone, who doesn't understand the performance and efficacy of Gemini 3.7 flash really 🙎
This is AT BEST a mid list with a lot of biases surrounding the choices. Don't look for OBJECTIVITY in it.
Pakistan: 171 & 135 in 3 days at Headingley.
That's 306 runs across 2 innings - England's top 4 alone made more in 1st innings.
The real number that matters: 37.3 overs to lose 10 wickets in the 2nd innings. Root's first Test back as full-time captain and the seam quartet made it look procedural. #PAKvENG #WTC25 #Cricket
Fragile isn't a mindset, it's a stat: 37.3 overs to fold in the 2nd innings.
No batter built a defensive base past the first hour. Compare to England's approach - leave, wait, punish width.
Pakistan's batters are still playing white-ball instincts on a Dukes ball. #PAKvENG #Headingley #Cricket
early thoughts on Grok 4.6 and Gemini 3.7 Flash
I'm surprisingly not enjoying Grok 4.6 as much as I liked 4.5. It's a great model, I just don't like it as much
It is definitely smarter, but it seems to use more tokens and turns complete slower. This makes it somewhat awkward to use. 4.5 was not as smart, but it's speed (and perhaps lack of thoroughness) kept me more in the loop which compensated for it's relative intelligence gap
I hope this changes with Grok 4.7 where it can simply hold its own on the merits of its intelligence vs other options
Gemini 3.7 Flash on the other hand has been pretty nice to use. It's very quick and it's writing style feels very "clean"
In omp, I've switched my default model to 3.7 Flash
MU: ATH $1,213 (June 25) -> $868 now, down ~29% while reporting record $41B+ quarterly revenue
WDC: ATH $800 (June 18) -> ~$438 now, down ~45% after beating on revenue AND EPS
Me: "surely earnings will fix the mood"
The market: "no."
This is not investing, it's a situationship. 💔📉 #MemoryStonks
Microsoft just dropped $90B revenue, +18% YoY, EPS $4.81 (beat estimates by 11%+ folks). Azure grew 43%. Cloud RPO up 84% to $678B.
Meanwhile stock is still down ~19% YTD. Market said "cool numbers, now show me the AGI ROI" 🧵
#MSFT
The plot twist nobody's memeing enough: Microsoft's Anthropic stake just handed them a $3.2B gain this quarter.
Translation: MSFT is now making bank on BOTH sides of the AI race - OpenAI exposure AND a Claude payout. Diversification, meet arbitrage.