Rong Ou

Podonos just raised $2.4 million in pre-seed funding, and the signal is loud and clear: the voice AI industry finally has someone measuring the sound, not just making the noise. Based in Los Gatos, Podonos is building the infrastructure layer that decides whether voice AI feels real, human, and ready for primetime. The industry is racing toward a projected $47.5 billion by 2034, but growth without standards is chaos. Podonos is making sure the hype actually holds up when you press play. At the center is ✦Soohyun Bae, PhD in Computer Engineering from Georgia Institute of Technology and Y Combinator W22 alumnus. His track record stretches from engineering roles at Google Maps to leading AR mapping at Niantic, plus co-founding Bobidi and TickTock AI. He has seen what happens when technology scales without proper #guardrails. With Podonos, he's betting the future of AI isn't just about how smart the models get, but how believable and trustworthy they sound. Podonos evaluates models across naturalness, similarity, emotion, recognition accuracy, pronunciation, tone, and resilience. Add in #personaconsistency, and suddenly you're not just testing machines, you're testing their ability to pass as human. The twist? They deliver results in under twelve hours, when the legacy approach drags on for two months. In AI time, that's the difference between relevance and irrelevance. Investors caught the beat. Serac Ventures, led by Kevin Moore, took the lead, backed by NAVER D2SF, the venture arm of South Korea's NAVER, and KAIST Venture Investment Holdings. Together, they've now backed Podonos twice, with this $2.4 million round stacking on top of a $750,000 raise in March 2025. Total funding sits at $3.15 million, and every dollar is pointed toward scaling #engineering, refining #analysistools, and launching into the U.S., Europe, and Southeast Asia. This isn't theory. Podonos already serves Resemble AI, Play AI, and Sanas AI, hitting six-digit ARR by November 2024 with just four employees. Their secret weapon is reach: 150,000 evaluators spread across nine languages and thirteen locales, forming a human-in-the-loop platform that blends speed, rigor, and global coverage. That reach lets Podonos deliver evaluations at a scale and pace that no competitor has matched. The strategy is ambitious: expand into high-demand sectors like #healthcare, #finance, #gaming, and #advertising, then move beyond voice into #multimodal model evaluation, spanning video AI and large language models. The mission isn't just to check if AI works, it's to decide if AI works for people. #Startups #StartupFunding #EarlyStage #VentureCapital #PreSeed #AI #AIInfrastructure #Voice #VoiceAI #VoiceTech #Infrastructure #Technology #Innovation #TechEcosystem #StartupEcosystem #Hiring #TechHiring If software engineering peace of mind is what you crave, Vention is your zen.

6

1 Comment

Are single GPU notebooks forcing you to downsample or don't let you experiment with large models? As datasets and models grow with the shift from structured to multimodal data, datasets and models no longer fit in the single GPU development environments. This blog highlights key considerations of distributed AI, how Ray lets you go multi-GPU without rewriting your code, and how Anyscale makes running it easy. https://lnkd.in/gWFYDuaY

21

Solid primer on the five architectures behind synthetic media. GANs, VAEs, Transformers, CNNs, RNNs. What's telling: the piece notes modern systems rarely use single architectures anymore. Diffusion models blend Transformer attention with U-Net structures. StyleGAN builds on progressive growing techniques from years ago. The toolbox keeps expanding. The combinations keep multiplying. #SyntheticMedia #GenerativeAI #AI #Deepfakes

Forge delivers 2.4–5.2× faster inference than torch.compile on NVIDIA B200 We benchmarked Forge against torch.compile(mode='max-autotune-no-cudagraphs') the strongest PyTorch baseline available across 7 production models Now publicly available with a free trial: https://lnkd.in/dnw_XmJH

35

2 Comments

👉️ https://lnkd.in/gAUt3gZB 🧠 Writing PyTorch extensions? Your memory allocator might be the hidden bottleneck. When you're building custom operations—or integrating specialized hardware—PyTorch's default memory management won't always cut it. This blog dives deep into how to write efficient, debuggable, and performance-tuned allocators for your extensions. Key insights include: 🚀 When (and why) to use custom allocators 🎯 Memory alignment, pooling, and fragmentation strategies 🔍 Debugging tools and sanity checks you wish you had sooner 🤝 How to coexist with PyTorch’s built-in allocator without reinventing the wheel 💬 Building custom extensions or facing memory headaches? Let’s trade strategies—what’s worked for you? #PyTorch #AIEngineering #DeepLearning #MemoryManagement #CUDA #PerformanceEngineering #PythonDev #MachineLearning

1

PyTorch + CUTLASS = 💜💚❤️ Huge applause to my friends on Meta’s PyTorch team for the incredible work integrating CUTLASS (NVIDIA’s open-source high-performance GEMM kernel generator) directly into the PyTorch compilation stack via TorchInductor. This is one of those things that quietly makes a *huge* difference: 💜 Epilogue fusion to inline ops like activation into the kernel → more compute, less memory hopping 💚 Auto-tuned, architecture-aware kernel generation at runtime with optimized block sizes and much more ❤️ Support for recent NVIDIA compute arch features such as TMA, tensor core FMAs, accumulator precision and so much more. And yes — all the fun but tricky bits like tile sizes, dtypes, and memory transfer params? PyTorch now handles that for you via smart integration with CUTLASS. Watch this episode of the PyTorch Compiler Series where Michael Lazos walks us through how to get maximum GPU performance from your PyTorch model *without* having to moonlight as a kernel engineer. NVIDIA GPUs offer amazing performance and efficiency to transform make your models go brrrrr with CUTLASS (or cuBLAS or Triton), and TorchInductor puts this power at the finger tips of every ML Engineer. So check out this PyTorch video and then give CUTLASS a spin with TorchInductor, and share how you use CUTLASS and NVIDIA GPUs to accelerate your workloads! #PyTorch #CUTLASS #AcceleratedAI #TorchInductor #GPUs #DeepLearning #Meta #NVIDIA

76

Anysphere, a three-year-old San Francisco startup that makes the AI-powered coding assistant Cursor, raised $900 million at a $9.9 billion valuation, reports Bloomberg, confirming a scoop by the Financial Times last month. The deal was led by Thrive Capital, with Andreessen Horowitz, Accel, and DST Global also opting in. https://lnkd.in/gtYGEPFu

4

On-device AI becomes powerful only when it works reliably on real hardware. In this Melange Walkthrough, Yeonseok Kim and Sun-Q Kim break down how the full on-device pipeline works, from model selection and quantization to real-device benchmarking and SDK deployment. Select. Benchmark. Deploy. This is how production-ready on-device AI should operate. Watch the full walkthrough below. #OnDeviceAI #EdgeAI #MobileAI #AIInfrastructure #ZETIC #Melange

26

Learn how to build an image search app using a series of guided code commits with Developer Advocate Guy Royse. He’ll guide you through generating image embeddings using freely available Hugging Face models and storing them in Redis for fast, scalable semantic search.  The workshop covers: ▶️ What embeddings are and how they work ▶️ How to generate image embeddings and store them in Redis ▶️ Building a semantic search app with Redis, Python, and JavaScript Get access here: https://lnkd.in/gbZ3imbz

44