Vipul Ved Prakash
Co-Founder & CEO, Together AI | Former Director of Engineering, Apple | Founder, Topsy & Cloudmark
Co-founder and CEO of Together AI, the premier cloud platform for training, fine-tuning, and serving open-source generative AI models at scale. Former Director of Engineering at Apple, where he led search and machine intelligence following Apple's $200M+ acquisition of his real-time social search company Topsy. Author of Razor anti-spam protocol and founder of Cloudmark. Deep systems engineering pioneer dedicated to democratizing compute and open-source model inference efficiency.
Areas of focus
Professional niches
Proof of Work
Together Inference Engine & FlashAttention-Optimized Model Serving
Engineered Together AI's custom inference kernel stack, implementing speculative decoding, custom FP8 FlashAttention kernels, and continuous batching to deliver the industry's fastest serving latency for Llama 3, DeepSeek, and Mixtral.
Highly optimized custom kernels require aggressive quantization and continuous hardware profiling for non-standard GPU architectures.
Context: Llama 3/3.1, Mixtral 8x22B, custom CUDA/Triton kernels, speculative sampling, continuous batching.
View missionRedPajama: Open Pretraining Datasets for Reproducible Foundation Models
Co-led the open-source release of RedPajama, a 1.2-trillion and 30-trillion token pretraining dataset replicating the LLaMA pretraining corpus to enable transparent, fully reproducible foundation model research.
Web-scale data deduplication and filtering at 30T scale require petabyte-scale distributed compute and risk residual benchmark contamination.
Context: 30T token web corpus, deduplication pipelines, MinHash LSH, toxicity filtering.
View mission