Asia2G Capital

Excited to Announce Our Latest Investment: Nomadic Drones We are very pleased to share that Asia2G Capital has added Nomadic Drones to our portfolio, further expanding our investments in the Smart Machine category. Nomadic Drones is transforming power infrastructure inspection with their innovative autonomous drones. What sets them apart is their proprietary inductive charging technology that allows drones to perch on power lines like birds, continuously recharging while conducting thorough inspections across vast infrastructure networks. The aging power grid across the US and Europe, combined with increasing electricity demands driven by AI adoption, has created an urgent need for more efficient inspection solutions. Traditional methods—manual piloted drones, helicopter surveys, or dock-based systems—all come with significant limitations in terms of cost, coverage, and efficiency. Founded by Andreas Moldskred (Norway) and Lauritz Weil (Germany), who met as exchange students at UC Berkeley, the team combines complementary expertise in software and hardware development. Their vision to revolutionize infrastructure inspection has already gained traction with three signed LOIs from European utility companies. While Asia2G Capital primarily focuses on Asian startups, including those from Korea, we occasionally make strategic investments in exceptional companies regardless of geography. Our investment thesis centers on finding vertical spaces where startups can win with minimal capital expenditure, particularly in industrial applications where smart machines can deliver great productivity gains. #VentureCapital #SmartMachines #EnergyInfrastructure #AutonomousDrones #Investing #NomadicDrones https://lnkd.in/et7WJtv3

Asia2G Capital recently hosted the A2G meetup, focusing on genAI / AI innovation in the enterprise realm. The event gathered over 30 VCs and industry leaders from prominent Silicon Valley companies. Archi KyoungRok Kong, the managing partner at Asia2G Capital, kicked off the event with a welcoming address, followed by Jihoon Jeong, the general partner at Asia2G Capital, who shared insights on CES 2025 from an AI perspective. Diving into the event, five speakers delved into various sectors: - Brian Pak from Theori discussed GenAI-based AI hackers versus World Class Human Hackers. - Jeff Baird from Celltrio presented the world's first end-to-end AI-based robotics for cell culturing. - Kevin Ha from CipherData explored how to revolutionize enterprise security operations with Gen AI. - Wonkyu Lee from ARGOS Identity highlighted innovations in identifying individuals and businesses in the gen AI era. - Subbu Rama from BalkanID delved into Gen AI-driven identity security, governance, and administration. The event also featured a dynamic panel discussion with esteemed founders and industry leaders, including Ken Kim from Kong Inc., Sheng Liang from Acorn Labs, Marco Palladino from Kong Inc., and James Barrese from Intuit. Special thanks to Kevin Kluge, Stephen Lee, Tae Hea Nahm, Jeff Brown, Sofi Choi, Hoshik Kim, Shyam Pillalamarri, Ram Motipally, Gurvesh Bhutiani, Masanori Yamato, Chris Yeh, Yogesh Badwe, Sangram Dash, Jay Kim, Jennifer Cho, Juye Kim, Perry Ha, Daniel Lee, David Kim, Jefferey Kim, Hyemin Kim and MinJoo Kim. #A2G #Asia2G

We're incredibly excited to see HyperAccel, a company we backed at the seed stage, achieve this significant milestone: a $40 million Series A funding round! This is a testament to their groundbreaking work in developing specialized AI semiconductors for LLMs. HyperAccel is developing an LLM Processing Unit (LPU), a dedicated hardware solution designed to dramatically accelerate LLM inference. Their technology directly addresses the limitations of current GPU-based solutions, offering a compelling alternative in the rapidly growing AI market. What makes HyperAccel stand out? Focus on LLM Inference: They're not just building another general-purpose AI chip. HyperAccel's LPU is purpose-built for the unique demands of LLM inference, a market poised for exponential growth. Cost and Energy Efficiency: Their LPU, leveraging low-power LPDDR5X memory (a world first!), promises significantly improved price-performance and power efficiency compared to traditional GPUs. In fact, they target 10x better price-efficiency in LLM inference when their ASIC chip is produced. Proven Execution: Despite being founded in early 2023, HyperAccel has already launched their LLM-specialized server, Orion, and successfully commercialized their LPU-IP, demonstrating their ability to execute rapidly. Strategic Partnerships: HyperAccel is already collaborating with major domestic and international corporations, validating the market's need for their technology. The Samsung 4nm process will be used for mass production, a partnership that underscores the quality and potential of HyperAccel's technology. Exceptional Team: Founded by Professor Jooyoung Kim from KAIST (and former Microsoft server hardware accelerator project lead), the team possesses deep expertise in AI hardware and software. This expertise is also why they were able to showcase the first domestic operation of Meta's Llama 3 in Korea. The team's ability to stay at the forefront of AI technology further strengthens our belief in HyperAccel's future. Global Ambition: With this funding, HyperAccel is poised to become a global leader in the LLM-specific AI semiconductor market. Investors are Vickers Venture Partners, Korea Development Bank, KB Investment, Mirae Asset Venture Investment, SBVA, Bon Angels, KDB Capital, LB Investment, Company K Partners. We were early believers in HyperAccel's vision to revolutionize LLM inference, and this Series A further validates their potential. We're proud to continue supporting them on their journey to reshape the AI landscape! Learn more about HyperAccel's innovative LPU technology: https://hyperaccel.ai/ #HyperAccel #LLM #LargeLanguageModels #AI #ArtificialIntelligence #Semiconductor #ASIC #LPU #SeriesA #Inference #HardwareAcceleration #LPDDR #A2G #Asia2G

HyperAccel, an AI semiconductor company founded in 2023, is a startup developing the 'LPU (LLM Processing Unit)', a semiconductor specialized for large-scale AI. The LPU is a semiconductor that optimizes memory bandwidth, crucial for large-scale generative AI computations, and integrates AI-specific logic. HyperAccel supports the advancement of generative AI technology and infrastructure development by accelerating inference in Large Language Models (LLMs). The company is challenging innovation in the AI semiconductor field across semiconductor chips, servers, and IP licensing. Asia2G Capital has made an investment in HyperAccel, recognizing the company's innovative approach to LLM inference. HyperAccel's strengths lie in their efficient and original IP, which offers cost-effectiveness while maintaining high performance. Furthermore, their product line includes FPGA server solutions capable of immediately executing a wide array of LLM models. The ongoing supply constraints faced by NVIDIA are expected to create significant opportunities for some AI semiconductor companies to experience substantial growth and demand. HyperAccel is well-positioned to emerge as a key player in this evolving landscape. This investment reflects Asia2G Capital's confidence in HyperAccel's potential to capitalize on the burgeoning market for AI acceleration hardware and establish itself as a formidable competitor in the industry. https://lnkd.in/ek43rf4X #LLM #AI #Semiconductor #LPU #HyperAccel

AI ?? ?? Top 5: 2024? 8? 3?? (Phi-3.5, #Midjourney, #Adobe, #Samsung) ??, Z?? 70%? AI? ??? ????? ???? ?? ??, AI ????? ???? ??? ?? ???, ??? AI ??? ?? ?? Magic Fixup ?? ???????, Phi-3.5 ?? ?? Midjourney, ????? ??? ?? ??. 25??? ??? ?? https://lnkd.in/gvRQwjHf

AI ?? ?? Top 5: 2024? 8? 2?? (#Grok-2, AI Scientist, #AMD, #Pixel, #Meta) X.ai, Grok-2 ?? ??. ??? ??? ???? ???? ? ?? Sakana.ai The AI Scientist ?? ??, AI? ???? Pixel 9 ??? ?? AMD, ?????? ??? ??? ZT ??? $4.9B ? ?? ??, ???? ????? ???? ?? ?? ?? https://lnkd.in/gPa7H-gf

AI ?? ?? Top 5: 2024? 8? 1?? John Schulman? Anthropic ??, Greg Brockman ??. OpenAI ???? ?? ?? ??, Character.AI ??? ??. ?? ? ???? OpenAI? ??? ????? ???? Figure.ai, Figure 02 ?? Groq, $2.8B ??? $640M ?? ??? D ???? ?? ???? ??, MS? ??? ?? https://lnkd.in/gbUK_6QQ

AI ?? ?? Top 5: 2024? 7? 4?? #OpenAI ? #SearchGPT ??, MS OpenAI? ???? ?? #AppleIntelligence, ?? #TPU ? ?? ?? #Llama 3.1 ?? (feat. ?? ?? ????? ???? ??) ???, #GenAI ????? 30%? 2025? PoC ?? ? ??? ??? ?? #Neura ????, ????? #4NE-1 ?? https://lnkd.in/gm36xGCx

AI ?? ?? Top 5: 2024? 7? 3?? #OpenAI, #GPT-4o mini ?? Andrej Karpathy , AI ?? ???? #EurekaLabs ?? ????, ?? ??? ???? Oxford Semantic Technologies ?? Menlo Ventures, #Anthropic? $100M ?? AI ?? ?? #Cohere, #Cisco, #AMD, #Fujits ????? $500M ?? ?? https://lnkd.in/gQdxX-GA

AI ?? ?? Top 5: 2024? 7? 2?? (#Amazon, #AMD, #Graphcore, #MobilityVLA) ???, ?? ????? AI #Rufus ?? ????? ?? ?????, Graphcore ?? ?? ???? Mobility VLA ?? ?? ???? ????, AI ??? ?? ?? AMD, ??? ?????? AI ??? #SiloLab ?? https://lnkd.in/gbiVwkJ5