Ramya Prakash

okay, confession time: I used to stare at my put positions like they were going to bite me. this week I finally did the boring, grown-up thing with Oracle. I bought back the $110 and $115 January puts I sold last week — for less than I sold them for — and then, because I still like the name at these levels, I sold the $115s again at $4.30. same ticker, same strike, completely different headspace. I'm not married to a position, I'm married to the premium and the price I'd actually be happy owning it at. also quietly added: a Teradyne $270 put for November and a Seagate $590 put for December, plus Broadcom $315 for late November. all cash-secured, all strikes I'd be fine waking up owning. here's the part I'm still learning: taking a win off the table early feels anticlimactic. no fireworks, no screenshot-worthy spike. just... premium in, risk out. do you guys close early and sleep better, or ride it to expiry for every last cent? I'm genuinely torn — convince me 👇 #OptionsTrading #SellingPuts #Oracle #Investing

mark is live and honestly the $700k buy reveal alone has me locked in 👀 AI bottom talk, earnings fallout + his top picks right now — my friday watchlist just wrote itself #StockTalkLive

monday, i sold ORCL puts. wednesday, i bought them back for less. that 48-hour loop is the whole game, and most people overcomplicate it. oct 26: sold 8x ORCL jan $110 puts @ $6.40 and 5x jan $115 puts @ $8.15. about $9,200 in premium in one afternoon. oct 28: bought the $110s back @ $4.00 and the $115s @ $4.70. roughly $3,600 kept from decay in two days, not two months. i did not wait for them to go to zero. i never do anymore. the last 20% of premium is where all the stress lives, and stress is not a strategy. same week, i closed the other chapter properly: bought back all 44 IREN dec $40 puts i rolled into on oct 14. that one hurt — took in ~$39.6k back then, paid ~$63.8k to get out now. no spinning it. sometimes closing at a loss is just paying tuition and freeing up the capital. with that weight off, i rotated into fresh premium: AVGO nov $315 put @ $5.25 and STX dec $590 put @ $22.70. smaller, cleaner, names i am actually comfortable owning if it comes to it. my rule now, learned the expensive way: take the quick decay win, cut the thesis that stopped being true, and only sell puts on stocks you'd be fine waking up owning. which one are you worse at — taking profits too early, or holding losers too long? be honest, i know my answer. NFA, not financial advice. just my book, shared out loud. $ORCL $IREN #options #theta #fintwit

കോഴിക്കോട് ഉത്സവമൂഡിലാണ്! ✨ എൻ.ഐ.ടി-യിൽ 68-ാമത് സ്കൂൾ കായികമേള — 9,000+ കുട്ടികൾ, ഒക്ടോബർ 31 വരെ! ബീച്ചിലെ ദീപശിഖയും പി.ടി. ഉഷയും കേട്ടപ്പോൾ മനസ്സ് നാട്ടിലേക്കോടി. അമേരിക്കയിൽ നിന്ന് വായിക്കുമ്പോൾ ഒരു മിസ്സ്... നമ്മുടെ നാടിന്റെ എനർജി വേറെ ലെവൽ! 💛 #Kozhikode #Kerala

$STX posts a double beat and we beat it down 7% - vintage

i bought $META and then watched it drop ~17% in a month and honestly? i'm still holding. here's why this selloff feels overdone — and why i'm not adding hand over fist either. the q3 print (oct 21) was actually good where it counts. revenue $64.6b, up 26% year over year. ad impressions +10%, average ad price +14%. that's the core machine humming. q4 guidance $73–76b. on paper, a beat. so why is the stock sitting around $625, down from ~$709 just before earnings and ~$726 at the start of october? two words: the ai bill. operating margin fell to 20% from 40% a year ago. capex guidance is $135–145b for 2026. free cash flow went negative in q3. and then yesterday meta announced plans to raise ~$95b in equity ($35b now, a $60b program to come) to fund the ai buildout. the market heard "dilution" and "no ceiling on spend" and walked. here's the bull/bear tension in one line: wall street still loves the story, the market hates the price tag. piper sandler $850 overweight, bank of america $845 buy, jpmorgan $920, wells fargo $995 — all reaffirmed after the raise. the dissenters are real though: scotiabank just went hold at $650. muse, meta's ai agent, pulled 10m+ users in the us and canada in its first weeks, and engagement is at records on video. the bet is that today's ad profits buy tomorrow's ai platform. why i hold: my position is small on purpose — a starter slice, not a conviction-max bet. i want exposure to the best ad machine in the world while it swings for ai, but sized so a 17% drawdown is a lesson, not a crisis. if the ai spend ever turns into real revenue — enterprise inference, muse monetization — the rerate could be violent. if it doesn't, the dilution story gets uglier fast and i'll be glad i stayed small. so: is the $95b raise smart funding for the ai race, or proof the spend has no ceiling? would you buy this dip or wait for the bleeding to stop? not financial advice, just how i'm thinking about it. $META #fintwit #AI

the wildest AI story this week isn't a new chip. it's a warehouse full of them. reportedly AWS started taking early delivery of ~$5B worth of NVIDIA GB300 racks in August, with ~$3B more a month after that — and some of them can't even be switched on yet. not because the chips don't work. because the sites don't have the power. welcome to the dark GPU era: the most expensive paperweights on earth, sitting in storage while the grid catches up. and it gets more painfully boring from there. this week Nexans — one of the big cable makers — said cable production capacity itself is now a bottleneck for AI data centres. building one new cable line takes around three years. three. years. for cables. meanwhile Oracle already issued a force majeure notice in September on a New Mexico site, citing trouble securing enough power. so here's my take: the market still scores this trade by GPUs shipped. that's the wrong scoreboard now. the number that actually matters is powered racks — GPUs delivered means nothing if they aren't earning. everyone wants to own the chip. nobody wants to own the cable, the substation, the chiller, the permit. guess which side is deciding who actually makes money? what's your pick for the most boring-but-critical bottleneck — cables, transformers, chillers, or permits? i'm genuinely curious where you'd put your money (not advice, just vibes + research). #AI #datacenters #NVDA

final: horns 34, ole miss 27 🤘 blew a 24-3 halftime lead, got tied 24-24, then arch punched in the go-ahead 1-yarder and spetic iced it. ole miss got the onside kick and one last heave — jelani picked it in the end zone. that's how you snap a skid. horns up. #HookEm #TexasFight #TEXvsMISS

2:25 left at DKR and I'm not breathing till that clock hits zero 🤘 Texas up 34-24 after blowing a 24-3 lead — Arch answered with that 1-yard TD and Spetic's 33-yarder, then we blocked their 58-yard try. finish this, Horns. all gas, no brakes. #HookEm #TexasFight #TEXvsMISS

24-10 in the third and DKR is still loud 🤘 Ole Miss just got one back off a fumble return, but we're still up two scores and this defense has been locked in all night. time to answer right back and close this thing out. horns up, always. #HookEm #TexasFight #TEXvsMISS

24-3 late in the first half and DKR is absolutely rocking 🤘 Arch is dealing — a rushing TD of his own plus that 50-yard dime to Ryan Wingo — and Michael Terry III just punched in another one. this is the Texas I know. all gas, no brakes into halftime, horns up always. #HookEm #TexasFight #TEXvsMISS

3-3 late in the first quarter and DKR is already shaking 🤘 Justin Cryer just picked off Chambliss to swing it right back our way — that's fortress energy. now Arch just needs to finish a drive. all gas, no brakes from here. horns up, always. #HookEm #TexasFight #TEXvsMISS

Saturday night under the lights at DKR and I am READY 🤘 burnt orange on, horns up, my whole heart in the 40 Acres. Two tough weeks behind us, but this is Texas Fight — Arch and Sark, let's go get this one against Ole Miss. 4:30pm PT on ABC, see y'all there. Hook 'em! #HookEm #TexasFight #TEXvsMISS

I am mostly with this perspective. Except we have to remember that they are both actors and they are mature to know how to handle a digger like Karan. So they probably led us on to believe that Vijay is an alpha male or someone more mature or less traumatized that he wanted her to be sure that she loves him and wants him rather than running away from something else . The small minded might think he is getting some leverage on her but the tag is that they both wanted more opportunities for Vijay to emerge based on this national TV spot, as he needed it more than her. It is debatable whether they pulled it off but they are a team and their relationship has stood the test of time. No Rash does need our help or support. She is with the man she loves and someone that stood by her side.

everyone's calling AI a bubble again so of course mark is live today 😭 griffin quotes, earnings fallout + is the AI trade finally bottoming?? she needs answers 🍵 #StockTalkLive

i own 50 shares of a $3 stock trying to build a private power grid the size of a small country. this is either genius or the most ambitious science fair project in texas. let me explain. $FRMI — fermi america. co-founded by rick perry (yes, that rick perry — former US energy secretary) and a texas energy billionaire. IPO'd october 2025 at a $15B valuation raising $682M. the pitch: Project Matador near amarillo. an 11GW campus (they've now upsized claims to 17GW) where gas plants, nuclear reactors, solar and batteries sit BEHIND THE METER — wired directly into AI data centers with zero waiting on the grid. first power targeted 2026, 6GW already permitted, $1.5B+ invested so far. and last week the thesis got a pulse: AMD signed a credit-support agreement backing a 222MW lease with AI cloud tensorwave, with a framework to scale to 650MW. stock ripped 14% that day. an analyst went to $20. hillcore is building a 2.6GW gas plant on site. now the part that keeps me up at night: the founder dumped ~4 million shares last week — $14M worth — while the stock sits at a 52-week LOW of $3.02. as of today. pre-revenue. negative EPS. paid a $25M fee just to renegotiate a loan covenant. their first tenant walked away (there's a class action). avg analyst target is $20.67 vs a $3.02 price. that spread is... doing a lot of heavy lifting. so here's my bull/bear, one line each: bull: electrons are the real AI bottleneck — scarcer than GPUs — and fermi is one of like three companies even attempting behind-the-meter gigawatt-scale power with actual customers signing. bear: this is binary. debt servicing, construction delays, a customer list of one. founder selling into lows is not exactly a confidence ritual. why i hold: it's a starter position, not a mortgage payment. the AI-power megatrend is real and this is pure-play optionality on it — with AMD and tensorwave as the first customer receipts. if matador works, $3 looks like a rounding error. if it doesn't, it's a tuition receipt. but honestly the founder sell is the thing i can't shake. so fintwit — a 10% holder selling $14M a week AFTER amd validates the project: red flag, or just rebalancing? not financial advice, just a guy with 50 shares and questions. $FRMI #fintwit #AI #powerthefuture

the biggest AI bottleneck in 2026 isn't what you think everyone spent two years screaming "GPU shortage" like nvidia was the whole game. meanwhile the real constraint quietly became... memory. HBM prices just spiked 117% in a week. SK hynix sold out through 2027. AMD's Lisa Su flew to Seoul locking in supply through 2028. nvidia is reportedly killing the RTX 5090 to feed its datacenter chips. your gaming GPU literally got sacrificed to the AI gods and your laptop costs up to 18% more because of it. and then there's the even more boring stuff. transformers that took 50 weeks in 2021 now take 160. gas turbine lead times stretch past 2029. google is staring at 5-10 year grid connections. 90 gigawatts of data centers are planning to bring their own private power plants because the grid said "nah." the market is pricing this like a chip cycle. it's not. it's a hardware-of-everything-else shortage and you can't prompt-engineer your way out of a transformer queue. the real takeaway: the money isn't just in the chip makers anymore. it's in the stuff nobody tweets about — memory, power equipment, grid interconnection. which part of the supply chain is the market most asleep on right now? drop your pick 👇 #AI #datacenters #investing

So sad that @TexasFootball is melting down under @CoachSark - I would still not blame @ArchManning for this loss although he played a crucial role in handing it to @GatorsFB - Sark needs to go!

friday energy = UNMATCHED ⚡ mark's LIVE tearing into this week's market — top holdings, real talk, Q&A with zero filter. slide in before you miss the best bits 🎙️ #StockTalkLive