Trajectory Ventures

Photonics: Light-speed computing for the AI revolution. By replacing electrical signals with light, photonics is poised to transform AI infrastructure by fundamentally reimagining how we process and move data. trajectoryventures.vc/ai-innovation/… #AI #Photonics

NIXXY (NASDAQ: $NIXX) at the 38th Annual @ROTHCapital Partners Investor Conference. Ritz-Carlton, Dana Point, California March 22-24th They’ll be meeting with investors and partners to discuss their growth across AI, fintech, and telecom infrastructure. globenewswire.com/news-release/2…

“Layla is flying today, with 1000s of flights and 100s of flight hours since 2022. We now have LOIs for more than 1,600 aircraft, $11B in customer interest, and expect to deliver aircraft and begin generating real revenue in 2026.” — @jdorris3 co-founder, CEO at @OdysAviation

Enjoyed talking with @robtfrank on @CNBC "Family Office Investor" about investing in #AI innovation and how artificial intelligence can either “augment humanity” or become a job-killing digital predator, depending on how Co's & regulators manage its growth cnbc.com/2023/05/08/ai-…

Introducing Bilt Hospitality for Restaurants. Bilt's newly launched hospitality platform is a fully managed concierge service that unifies the guest experience before, during, and after every visit. We're proud to have some of the greatest names in food and hospitality with us.

These Two Startups Are Teaming Up to Prevent a ‘Pearl Harbor Moment’ in Space. @PortalSpaceSys and @paladinspace_ are partnering to launch a new service in low-Earth orbit to remove the growing dangers of space debris. Here’s how it will work. inc.com/chloe-aiello/t…

On @BloombergTV our CEO @PatrickTBowen explains why inference is fundamentally power-limited — and why solving it at the physics level unlocks 100× gains in both speed and energy efficiency.

Neurophos is transforming AI with their optical compute. Using meta materials that enable light based AI infrastructure to exponentially advance the bandwidth of data centers and speed artificial intelligence. While reducing energy consumption. spectrum.ieee.org/optical-metama… #AI

Andromeda flight engine awaiting shipment to our test site in Moses Lake, WA. 🔥 Big shout out to the local community and to our partners in the area like the @PORTOFMOSESLAKE and Grant County Sheriff's Office for being good neighbors and sharing the excitement as we work relentlessly toward launch.

$MSFT employee on why $GOOGL has the most integrated stack among hyperscalers ( $AMZN ): - The expert believes enterprise AI usage will naturally split across multiple models rather than consolidate around a single one. Simpler tasks are well served by smaller open-source models, making it hard to justify the cost of a frontier model. On more complex or sensitive applications, customers will gravitate toward leading models, but switching between them is not straightforward. This creates natural stickiness, particularly for applications where significant testing and validation have already been done. - The expert emphasizes that as newer applications emerge across the stack, entirely new budget categories are being created to support them. On the SaaS side, he expects significant disruption. As foundation models improve, specialized software applications face real pricing pressure, with many likely to be rebuilt directly on top of foundation models in customized PaaS environments rather than purchased as off-the-shelf solutions. - According to the expert, $GOOGL is the best-positioned hyperscaler over a five to ten-year horizon. The combination of proprietary hardware like TPUs, frontier models like Gemini, consumer data at scale, and the pace of acceleration they have shown in recent years gives them a uniquely integrated stack across software, hardware, and data. - He sees $MSFT as strong, particularly given the OpenAI partnership and their platform strategy of supporting any model at scale, though he notes OpenAI is no longer exclusive to $MSFT, which changes the dynamic somewhat. $AMZN he views as catching up, with strong foundational cloud infrastructure and talent, and the recent OpenAI investment signaling a more aggressive push that could meaningfully close the gap with $MSFT. - The expert describes the relationship between hyperscalers and model providers as deeply entangled, friends and enemies at the same time. $MSFT benefits from OpenAI's models driving cloud revenue while OpenAI competes directly with $MSFT's own consumer and enterprise products. The expanding market has kept this tension manageable so far. His view is that the ultimate winner will be whoever can offer the best model at the lowest cost with the most complete integrated stack, covering models, agent frameworks, and cloud infrastructure in one offering.

NIXXY Signs Infrastructure Services Contract with Telforge, Inc. to Manage an Estimated $60,000,000 in Telecommunications Traffic. Click the link below to read more: nasdaq.com/press-release/…

$NVDA just put $4B into $LITE and $COHR. $AVGO is shipping CPO switches. $MRVL demoed liquid-cooled CPO at OCP. Everyone's talking about who builds the optical engine. Nobody's asking the harder question: who supplies the light? Silicon can't make a laser. Every SiPho chip needs one from outside. Three microcaps are trying to be that supplier. All three surging. All three very different. $SIVEF (+63% in 5 days): InP DFB lasers. Partners with Ayar Labs, POET, imec. Closest to real revenue. QD Laser (6613.T, +33%): Quantum dot physics are real. But their actual customers are in biomedical, not datacenter. $ALMU ($254M mcap, 14 employees): Growing III-V directly on silicon. I visited their booth at Photonics West. Asked about yield. No clear answer. I wrote a full deep dive. Architecture from Gen 1 (EML) to Gen 4 (monolithic). Cost breakdowns. Financials. Bull/bear for each. First half is free!!! Link: "photoncap.substack.com/p/the-silicon-…" h/t @Semicon_player for the suggestion to dig into this one.🫡

We’re advancing Passage™ CPO & Guide® VLSP™ Lasers while scaling for HVM. From live demos @ofcconference to building an open, interoperable ecosystem with industry leaders, we’re delivering the AI infrastructure the industry demands.

@aleabitoreddit Want to put this one in your radar screen $NIXX.

AI infrastructure company @Nixxy $NIXX, which is advancing telecom and fintech convergence, announced it has signed a $60,000,00 infrastructure agreement with Telforge, a wholly owned subsidiary of FingerMotion (NASDAQ: $FNGR). trajectoryventures.vc/ai-innovation/… #AI $NVDA $NBIS

Inside @LightmatterCo's wide and parallel silicon photonics wins, which are driving the next phase of AI infrastructure performance. trajectoryventures.vc/ai-innovation/… #AI #Photonics | @TrajectoryVC

FraudNet enabled a leading mobile payment network to modernize its fraud and risk infrastructure, combining real-time monitoring, sanctions screening, machine learning, and more to improve operational efficiency and support multiple products and use cases across its network. Learn how: buff.ly/zrOtWE5
 #transactionmonitoring #policymonitoring #fraudprevention #frauddetection #ai #ainative #machinelearning #merchantrisk #merchantriskmanagement #merchantscreening #payments #acquirers #paymentprocessor #portfolio

Inside @LightmatterCo's wide & parallel silicon photonics wins, which are driving the next phase of AI infrastructure performance. This is the beginning of the most important and biggest infrastructure shift, enabling AGI & reducing energy consumption. trajectoryventures.vc/ai-innovation/… #AI

The upcoming CPO / Silicon Photonics Bottleneck Cheat Sheet: $SIVE, Sumitomo, $LITE, $COHR, $AVGO, $MTSI, $AAOI - Light Source (CW DFB Lasers) $TSEM, $GFS, $UMC, $TSM, $INTC - SiPh foundry $NOK, $CIEN, $CSCO, $COHR - DCO $HIMX, FOCI (3363.TWO) - Micro-lens + Fiber Arrays $POET - Optical Interposers $SOI, $AXTI, Shin-Etsu - Substrates $FN, $ASX, Innolight, Eoptolink - Optical Packaging and Assembly $MTSI, $SMTC, $MRVL, $MXL - Analog/Mixed-Signal ICs $LWLG - Speculative Modulator Materials. $GLW, $APH, $TEL, $FIT, Fujikura - Connectors and Fibers $FORM, $KEYS, $VIAV, $AEHR- Test & Measurement $BESI, $SMHN, $ONTO, $CAMT - Advanced Packaging & Hybrid Bonding Many are private companies from Lightmatter, Ayar, Ranovus and others. Now... Everyone is asking... How do you profit? If you look at the forecast for CPO TAM, it's a straight line up, and next year is inflection point for CPO mass deployment. The alpha is capturing the rotation: From the current EML bottlenecks ( $LITE, $COHR type) to SiPh / CW DFB architectural winners for CPO. Highest upside potential are the ones that aren't included in current cycles. But that are in the next. Companies like $SOI, $SIVE, or $AEHR are perfect examples. Ride the current pluggable bottleneck like $AAOI. But the alpha is frontrunning institutions with the next CPO bottleneck. The capital rotation is inevitable.

@aleabitoreddit @TrajectoryVC Looking forward to the cycle of light 📈🚀 it’s finally here like a light switch was turned on a few weeks ago. I have been track #photonic conversations on Twitter and LinkedIn, and it’s lit up in the last 4 weeks.

Photonics innovator @LightmatterCo achieves record 1.6 Tbps per fiber to accelerate AI Optical Interconnect. Their strategic @Qualcomm collaboration combines high-performance connectivity w/ Lightmatter’s photonic engine to achieve an 8X bandwidth-density. trajectoryventures.vc/ai-innovation/…