NuMind

Tensil automates machine learning hardware accelerator design. Our goal is to enable ML in every device. We've developed an algorithm that can design the best hardware accelerator for your machine learning model, unlocking 10x better performance per watt-dollar than GPUs and other hardware accelerators. Our technology enables our customers to create unbeatable products with Tensil custom hardware at the core.

shared extracted concept terms: model, machine; shared product terms: create, custom, model, best; same category: AI infrastructure; shared affiliation: Y Combinator

InferEdge builds agentic AI for the buy-side - systems that run autonomously across research, portfolio management, and risk, for equity and fixed income. The point isn't generating a generic note - anyone can do that now. For example, when earnings season hits, our agents produce the quarter's coverage in your analysts' own tone and format, matching how your desk already writes, so it doesn't need a rewrite before it's usable. We build on top of LLMs, turning raw model capability into consisten

shared extracted concept terms: model; shared product terms: research, example, when, analyst; shared affiliation: Y Combinator

AI-assisted virtual mental health clinic, leveraging deep-learning, predictive analytics and large language models to provide advanced intake matching with the best therapist, measure care progress throughout therapy, and support clients between sessions and beyond therapy

shared extracted concept terms: model; shared product terms: leverag, large, language, model; same category: AI infrastructure

We build Voice Models for regional languages (like Urdu, Bengali, Greek etc) and make them available through simple APIs and SDKs. Developers use us to create Voice-first digital experiences (eg. Voice eCommerce, Voice Banking, Voice AI Tutors etc). We gather our own data, develop our own labelling tools, train our own models, and design our own voice infra. This fully integrated approach enables us to offer the best models and the best developer experience for regional languages. --- People who

shared product terms: create, model, offer, best; same category: AI infrastructure; shared affiliation: Y Combinator

Interfaze is an AI model built on a new architecture that merges specialized DNN/CNN models with transformers for tasks that require deterministic output and high consistency like Document intelligence, Object detection, Classification, Audio understanding and more. Interfaze outperforms large lab models like Gemini flash, Claude Sonnet and the GPT-5.4 mini series Architecture paper: https://arxiv.org/abs/2602.04101 Full open benchmarks: https://interfaze.ai/leaderboards

shared extracted concept terms: gpt, model; shared product terms: model, large, gpt, org; same category: AI infrastructure; shared affiliation: Y Combinator

In Mexico and US trucking demand outpaces capacity, yet 40% of the time trucks roll empty. BeGo uses Machine Learning to reduce empty returns, deadheads, and carbon emissions. BeGo is co-founded by two brothers, Jasiel, with 10+ years of experience in the industry, and Ivan, ex-Tesla engineer with experience in product design and machine learning.

shared extracted concepts: machine learning; shared product terms: uses, machine, co, found; same category: AI infrastructure; shared affiliation: Y Combinator