Wealth Effect

@DEBU_NEOGI I am being plain honest , no sugar coating. My level of thinking and operation has changed entirely in te lsat 2-3 years since i have known the story of Debu da !!

In 1996, I earned ₹8,000/month. In 2026, my stock portfolio is worth ₹80+ crores. But between these two numbers…   I almost went bankrupt. In 1999, I entered the stock market. Not with knowledge.   With tips. Not with savings.   With loans. I borrowed ₹14 lakhs   (3× my annual income)   at 19% interest from 3 banks. I thought I was smart. Then the dot-com crash happened. Markets fell 50–60%.   My portfolio fell 90%+. Those stocks never came back.   They were operator-driven.  I just didn’t know it then. I was 28. Newly married.   One child.   Father retired.   House under construction.  And 3 bank loans at 19%. I was a qualified CA.   And completely broke. That was the loneliest year of my life. That phase gave me one rule I never broke again: Never invest borrowed money in something you cannot independently verify. I stopped taking tips.   I stopped borrowing to invest. I started: * Reading annual reports   * Meeting promoters   * Attending AGMs  27 years later,   banks offer me large credit lines. I haven’t used a single rupee of it for investing. Because the scar of 1999 never left me. And that scar built everything I have today. Your process will always matter more than your outcome. What was your biggest money mistake? Most people want multibagger stocks. I built wealth by avoiding blow-ups.

@ghostrider1159 @michaelsikand old news

@michaelsikand What about $SIVEF ?

$LITE is down today. It's not because the smart money is rotating out of photonics. It's rotating overseas where the mispricing is. Here are 3 European AI stocks I own that still trade at a fraction of their peers, creating asymmetric upside. 1. Aixtron ($AIXA / $AIXXF OTC) 🇩🇪 Owns 90%+ share of the MOCVD reactors that grow the indium phosphide crystals used to make AI data center lasers. Every laser sitting inside an Nvidia or Broadcom optical transceiver started inside one of these machines. Customers: Coherent, Lumentum, Nokia/Infinera, SMART Photonics. Lead times for the lasers themselves are now stretched past 2027. It's already re-rated significantly but still trades at 38x EV/EBITDA versus 84x for comps like $BESI. Same critical work, less than half the multiple. 2. LPKF Laser ($LPK / $LPKFF OTC) 🇩🇪 Not a photonics stock per se, but glass substrates are the next generation of AI chip packaging: Intel committed to high-volume by 2030, Samsung is piloting, TSMC is evaluating also a bet on the future architecture of semis. LPK owns the only working laser process (LIDE) that drills precise holes in glass without cracking it. 293M EUR market cap, pre-rev inflection, no meaningful EBITDA yet. That's the whole point. You're paying a small option price for a binary glass wins outcome where 4 to 10x is on the table. Soitec ($SOI / $SLOIY ADR) 🇫🇷 You've heard me screaming about this for a few months. The Photonics-SOI wafer that TSMC's COUPE and GlobalFoundries' silicon photonics platforms are built on top of. If a chip from Nvidia, Broadcom, Marvell, or Cisco contains silicon photonics, the starting wafer is probably SOI's. Trades at 15x EV/EBITDA versus 84x for US comps. Management is targeting 2 billion dollars in revenue at 40% EBITDA margins (~800M EBITDA). The stock got punished for unrelated RF-SOI smartphone cyclicality while their AI revenue grows underneath. Crown jewel of these three in my opinion and my largest position respectively.

@michaelsikand Dont forget XFAB & GHH.L

@aleabitoreddit Add XFAB to your list :) Oh i forgot to mention GHH.L

Glad to hear it! I've went long and wrote thesis posts on about out 15 different stocks that hit 100-1000%+ YTD? 1. $AXTI 2. $AAOI 3. $SIVEF 4. $LITE 5. $IQE 6. $AEHR 7. $CRCL 8. $EWY 9. Unimicron 10. Nitto Boseki 11. $OSS 12. $GDRZF 13. $RPI 14. $SOI 15. $ALRIB Not including others like $TSEM that are about to hit triple digit returns too in a month. The amount of hate people like myself get for posting free ideas over the internet is pretty insane TBH. Starting to make sense why people just set up $20,000 paywalls and sell info to Western institutions instead of helping out salty retail investors (especially over in Europe). But helps me keep motivated to keep posting with these positive comments.

IMO the most undervalued SI Photonics company - XFAB !!

$SIVE / $SIVEF – Sivers Semiconductors Product Flow 🧵 A tiny Swedish semiconductor company. Here's where their chips actually end up. It's so many that I probably missed a few. Please let me know. 👇 ━━━━━━━━━━━━━━━━━━━━━━ 🏭 WHAT THEY MAKE: 🔵 PHOTONICS – InP DFB Lasers & Optical Amplifiers 🔴 WIRELESS – mmWave Beamforming ICs 🏴󠁧󠁢󠁳󠁣󠁴󠁿 Own InP fab in Glasgow (Photonics – primary) 🇺🇸🇩🇪 GlobalFoundries USA/Germany (Wireless) 🇹🇼 WIN Semiconductors Taiwan (Photonics – volume) ━━━━━━━━━━━━━━━━━━━━━━ 🤖 AI DATACENTERS → AYAR LABS Optical I/O chiplets inside NVIDIA GPUs → Microsoft / Google / Amazon / Meta → POET Technologies Optical Interposer modules → Celestial AI → Marvell ecosystem→ Hyperscalers / AI infrastructure → O-NET + ENABLANCE External Light Source (ELS) modules → AI datacenters & HPC → JABIL Optical module assembly → Hyperscalers ━━━━━━━━━━━━━━━━━━━━━━ 🍎 BIOSENSING / CONSUMER HEALTH → US FORTUNE 100 CUSTOMER (anonymous) 5+ year relationship \$17M+ invested in development Advanced optical sensing applications 50,000+ laser devices delivered "Getting closer to volume phase" Rumored: Blood glucose monitoring in next-gen wearables ━━━━━━━━━━━━━━━━━━━━━━ 🚗 AUTONOMOUS VEHICLES → AEVA FMCW 4D LiDAR (Sivers tech on all platforms) → NVIDIA DRIVE HYPERION ✅ Volvo Trucks ✅ Stellantis ✅ Top European Passenger OEM ✅ Robotics & Factory Automation + All OEMs selecting Hyperion Start of Production: 2028 ━━━━━━━━━━━━━━━━━━━━━━ 🛡️ DEFENSE & ELECTRONIC WARFARE → BAE Systems + MIT Lincoln Lab + Columbia Univ STAR Full-Duplex Arrays → US Navy / US DoD → Raytheon + Ericsson Next-gen 5G/6G RF chips → US DoD + Global 5G → Arena Physica → Anduril Industries → US Defense systems ━━━━━━━━━━━━━━━━━━━━━━ 🛰️ SATELLITE COMMUNICATIONS → @allspaceltd (UK) Electronically Steered Antenna terminals Certified on SES O3b mPOWER ✅ US DoD (Programs of Record) ✅ UK Ministry of Defence ✅ Allied & Coalition defense ✅ Aerospace / Maritime / Rail → DOOSAN (Korea) Ka-band ESA panels → Korea SATCOM infrastructure → THORIUM SPACE (Poland) Multi-band mmWave chips → Ground terminals + Space deployment ━━━━━━━━━━━━━━━━━━━━━━ 📶 5G & FIXED WIRELESS ACCESS → TIER-1 Next-gen beamforming transceiver → Global telecom operators → TACHYON NETWORKS 28GHz antenna modules → Fixed wireless broadband operators ━━━━━━━━━━━━━━━━━━━━━━ 🎯 CONFIRMED END CUSTOMERS: 🤖 AI → Microsoft, Google, Amazon, Meta 🍎 HEALTH → Fortune 100 (rumored: Apple Watch) 🚗 AUTO → Volvo, Stellantis + Hyperion OEMs 🛡️ DEFENSE → US DoD, UK MoD, Anduril 🛰️ SATCOM → SES O3b mPOWER, Korea, Poland 📶 5G/FWA → Global telecom operators ━━━━━━━━━━━━━━━━━━━━━━ 6 megatrends. 1 supplier. $SIVE $SIVEF #Semiconductors #AI #Defense #Photonics #mmWave #5G #SATCOM

Sivers is transitioning from an engineering-driven business to a scalable, product-led semiconductor company. 📈 Pipeline growing (+64%) ⚙️ Product mix improving (+90%) 🚗 LiDAR ramp coming 🧠 AI datacenter exposure 💰 Model pointing to high margins Not financial advice.

$SIVE (Sivers Semiconductors) - This could be a MASSIVE player in the photonics industry for decades. As I connect the dots, it gets very interesting & exciting. Let me tell you why. Sivers Semiconductors manufactures high-power laser arrays (the critical light source) for its partners. Now who do they partner with? 1) Ayar Labs is one of them. They integrate these lasers into Optical I/O Modules (SuperNova™ & TeraPHY™) to build optical engines. Hyperscalers like $AMZN and $MSFT are actively preparing to deploy these modules and engines to interconnect their AI accelerators (Trainium/Maia) in massive data center clusters for AWS and Azure workloads, particularly for high-efficiency inference. This confirms $SIVE's role in the Amazon & Microsoft supply chain. 2) Sivers lasers are also supplied to $POET Technologies, which has a long-standing collaboration supplying optical engines to Celestial AI. $MRVL acquired Celestial AI in a ~$3.25B deal, and it expects Celestial AI to ramp to $500M-$1B annual revenue run-rate by FY2028-2029, focused on AI optical interconnects and co-packaged optics (CPO). $SIVE could be the “secret leader in InP lasers” sitting behind POET → Celestial AI → Marvell, positioning them as a potential upstream enabler for Marvell’s AI accelerator/XPU architectures that use advanced optics. The end-users in the chain could potentially include $META, $GOOGL (reported to be among the hyperscalers exploring Celestial AI’s Photonic Fabric), $MSFT, and $AMZN. 3) Ayar Labs has received investments from $AMD, $INTC, and $NVDA. All three are closely working with Ayar Labs to integrate its photonic technology into high-volume manufacturing processes and replace traditional electrical interconnects in high-performance AI data centers. $SIVE partnered with O-Net and Enablence (March 2026) to build external light sources for co-packaged optics (CPO) - the tech $NVDA needs for efficient scaling in post-Vera Rubin systems. Today, the main suppliers for NVIDIA’s CPO/optical needs are Lumentum $LITE and Coherent $COHR, secured by NVIDIA’s huge commitments ($2B investment each + multibillion-dollar purchases, March 2026) for reliability. But in the same month $NVDA gave $2 billion to each of those companies, it also co-led a $500M Series E in Ayar Labs. This could be that NVIDIA is betting on both horses: Lumentum/Coherent for immediate massive scale-out, and Ayar/Sivers for potential next-generation in-package integration. NVIDIA’s Chief Platform Architect has noted that Ayar’s optical I/O is the foundation for the “next million-X” speedup in AI. 4) $SIVE is also a Promoter Member (leader) of the CW-WDM MSA, the group defining the laser standards for AI optics. Broadcom is an Observer Member of this same group. They are literally writing the rulebook together for how lasers (Sivers) talk to switches (Broadcom). 5) Another thing is that demand massively outstrips supply today. AI-driven optical transceiver shipments are exploding (~24M units in 2025 to ~63M in 2026), causing severe InP laser/substrate shortages, lead times beyond 2027, and booked orders through 2026. Even with Lumentum/Coherent ramps, bottlenecks persist in epitaxy and high-power lasers - NVIDIA’s deals secure their needs but don’t fix the industry-wide crunch. The future possibility and current ongoing demand for these lasers could put $SIVE in a multi-decade supply chain across the ecosystem of the biggest AI companies in the world. That boosts my conviction big time! I hold a 5% portfolio position in $SIVE, since it is a BUY for me. NFA.

.., we're looking at ~10.6B SEK market cap (~34 SEK/share). Not Inclu $SIVE forward sales growth potential, 3–4x in this cycle. Misunderstanding of this potential is the actual disconnect in media and the true reason behind the recent $SIVE correction after forced selling etc.

On top of this, self driving car sales volumes has been shared from major US customer, as been completely missed in Media, this is major breaking NEWS for $SIVE. Production from Q4 2026. sivers-semiconductors.com/press/strategi…

Sivers Semiconductors ( $SIVE / $SIVEF) – Fundamental milestones vs. market noise. While the stock is seeing short-term volatility, the fundamental case has strengthened more in the last few weeks than in the past several years. Here are three massive catalysts the market is currently overlooking: The 1.6T Validation: Information and displays at OFC 2026 confirm that Sivers’ InP laser technology is a key component in the new 1.6T (Terabit) ecosystem. With Jabil ($JBL) debuting their 1.6T LRO modules, Sivers is now officially positioned as a critical high-power laser source for the next-gen AI infrastructure. EU Validation & Funding: Sivers is a confirmed partner in the E2PackMan Horizon Europe project (ID: 101194187). This consortium, with a total budget exceeding €90M, focuses on advanced chiplet packaging. This provides critical non-dilutive funding and puts Sivers in the same room as the industry’s biggest players. Technical Efficiency (pJ/Bit): Official data from OFC confirms that InP-based solutions (Sivers) deliver 3-4 pJ/Bit, compared to 4-6 pJ/Bit for traditional Silicon Photonics. In an AI data center where power density is the #1 bottleneck, a 30% energy saving is the difference between a scalable product and a cooling nightmare. With the Bootstrap Europe refinancing secured and the LiDAR ramp ($53M-$138M potential) confirmed for 2026, the technical and financial foundation is firmer than ever. Volatility is the price you pay for being early in a supercycle. Fundamentals eventually win. 📈 #Sivers #AI #Semiconductors #OFC2026 $SIVE $SIVEF $JBL $POET $LITE $COHR

Today I was called a pumper and an unintelligent investor. $LITE $CIEN $AAOI I want to take a second to reflect on that. First, I absolutely am not above criticism. I am learning in public, and as my platform grows, I take that responsibility more seriously every day. There are absolutely ways I can improve, especially in how I frame risk and how I discuss smaller or earlier-stage companies. That is something I have already been actively working on. But I also think it is important to be fair. Most of my public focus over the last six months has been photonics, with some energy and defense around the edges. Those have been very strong areas of the market. I do not say that to boast. I almost never talk about gains. I say it because if the main thing someone knows about me is that I have been concentrated in one of the best-performing sectors in that period, calling me unintelligent does not feel like a serious critique. As for the “pumper” label, the companies mentioned by this user were names like Lumentum, Ciena, and Applied Optoelectronics. These are not stocks that move because of my posts. Lumentum and Ciena are massive, widely covered companies. AAOI is smaller, and I am very bullish on it, but I have also repeatedly talked about the risks, the execution requirements, and the downside if things do not go right. Most of what I share comes directly from public company materials. My goal is usually to connect dots, simplify the story, and make a very confusing space easier to understand. I am not trying to invent narratives that will never come to fruition. I am trying to make information more digestible. I have also become more selective over time about what I discuss publicly. There are smaller and more speculative companies I find interesting, but I have intentionally reduced how much I talk about their upside because I do not want people taking positions just because I am excited about something or think that the technology is really interesting. That feels irresponsible. So if the feedback is that I should continue to be careful, disciplined, and thoughtful with how I communicate, I agree. I want that too. But I do not think being openly bullish on large companies that are executing well in a strong sector automatically makes someone a pumper. And I do not think disagreement alone makes someone unintelligent. I am always open to different views. I welcome pushback. I welcome better analysis. That is how we all get sharper. I just hope the conversation stays centered on the work, the facts, and the ideas. My goal with my platform moving forward is increasingly to question things. Stress test ideas. Second guess theses. Because the music will stop playing on the sector eventually. And I want to make sure that I, and those that follow me, are the most informed as possible to find the best and most resilient companies out there. Have a great day!

The photonics trade seems overcrowded at this point 6-month performance: Corning | $GLW: +78% Lumentum | $LITE: +425% Lightwave Logic | $LWLG: +75% Fabrinet | $FN: +64% Ciena | $CIEN: +210% Nokia | $NOK: +74% Applied Optoelectronics | $AAOI: +327% Coherent | $COHR: +156%

$SIVE has gotta be the highest upside stock I’ve seen in this market since $AXTI? No way markets missed the CW laser light source for Jabil, Marvell (Celestial via $POET), O-Net, Ayar ( $NVDA, Mediatek backed)… At a $140M valuation. ($350m now) Not only do you get the most direct laser exposure to future CPO scale up? But also this cycle’s 1.6T pluggables with $JBL (formerly Intel Silicon Photonics division) coming soon. With Win Semi bridge capacity scaling needed for hyperscaler supply chains. Don’t think 99.9% of people realized the sheer scale of this yet.

Sivers is not a company born overnight. They have been preparing for this day $SIVE youtube.com/watch?v=gWct-9…

Everyone is talking about $SIVE / $SIVEF in the networking layer (Jabil, POET), but FinTwit is completely missing Sivers’ most explosive catalyst: The GPU Heat Problem and "Optical I/O". 🧵👇 If you are tracking the physical limits of Nvidia's NVLink and the copper bottleneck, you know that GPU-to-GPU communication must move to light. But there is a massive physics problem that very few investors understand. 1. The 1000W GPU Heat Problem 🔥 Next-gen AI chips run incredibly hot. Lasers are notoriously sensitive to heat; if an Indium Phosphide (InP) laser gets too hot, its wavelength drifts and the data transmission fails. Therefore, you cannot put the laser chip near the GPU. 2. The Solution: "The Power Supply of Light" (ELS) 🔦 To solve this, the industry is moving to Co-Packaged Optics (CPO) utilizing an External Light Source (ELS). Instead of putting the laser on the chip, you put a "dumb" continuous laser far away in a cool part of the server rack, and pump raw, unmodulated light via fiber into the silicon photonics chip sitting next to the GPU. 3. Sivers’ Hidden Moat: High-Power CW Arrays ⚡ This is where Sivers ( $SIVE) holds a massive, under-discussed advantage. They specialize in High-Power Continuous Wave (CW) DFB laser arrays. They are not just building components; Sivers is part of the elite CW-WDM MSA standard group—sitting at the same table as Broadcom, Intel, Arista, and Nvidia—helping to standardize these exact multi-wavelength laser sources for next-gen AI compute. 4. The Lightmatter / Ayar Labs Connection 🌐Private unicorns like Ayar Labs and Lightmatter are currently valued at billions for designing these exact Optical I/O architectures. But they are fabless. They need a physical InP foundry capable of churning out flawless, high-power CW laser arrays to act as the "engine" for their systems. Sivers is one of the very few independent foundries in the world capable of supplying this. 5. The Ultimate Tell: Skin in the Game 👔 Deep tech can be complex, but insider buying is a universal language. CEO Vickram Vathulya already holds a massive stake in the company (over 3.4 million shares) and was just buying even more shares on the open market this very month (March 2026). When the CEO of an InP foundry is loading up his personal portfolio right before the hardware industry transitions to Optical I/O, you pay attention. The Takeaway ⚖️ Market participants are pricing Sivers purely as a telecom/networking component supplier. They haven't realized that Sivers’ InP100 platform is uniquely positioned to supply the external light sources required to save the GPU compute layer from melting. A ~$325M market cap for a de-risked foundry sitting in the middle of the Optical I/O revolution is a generational mismatch. $SIVE $NVDA $AVGO #Semiconductors #OpticalIO #AIHardware #TechStocks

@MarketMaestro1 And this is a post written by a person with very good insight.

@MarketMaestro1 A picture says more than 1000 words