Top Deep Learning Startups in 2026

Observe.AI, a leader in Contact Center AI, transforms customer experiences and improves agent performance by helping top brands analyze 100% of interactions, extract actionable sentiment and interaction insights, and streamline workflows. With Observe.AI, businesses transcribe every interaction with high accuracy and coach agents while gaining full visibility into their customer service operations. Observe.AI brings the power of agent assistance, automatic speech recognition, and Natural Language Processing (NLP) to modern contact centers and their frontline teams. Observe.AI is trusted by more than 160 customers and partners, including National Debt Relief, Root Insurance, Alcon Laboratories, and Pearson. Backed by Menlo Ventures, Next47, NGP Capital, Emergent Ventures, Scale Ventures, Nexus Ventures, and Y-Combinator, Observe.AI’s headquarters is in San Francisco with an office in Bangalore, India. For more information, visit observe.ai.

We are building the next generation of voice AI and LLM platforms: - It is on-device - It runs anywhere, from Arduino to web browsers - It beats FAANG in accuracy, latency, and reliability

Hey, we’re Deepgram! We’re an AI speech platform whose mission is to make every voice heard and understood. Our end-to-end deep neural network is revolutionizing the speech-to-text market and taking on the big guys. We’re redefining what companies can do with voice by offering a platform with AI architectural advantage, not legacy tech retrofitted with AI. Deepgram delivers highly accurate, actually usable transcription, right out of the box – or you can train a model for even greater accuracy on the unique words that matter to you. ✅ Live-stream or pre-recorded ✅ On-prem or in the cloud ✅ Scalable and affordable ✅ Dozens of languages and use case models ✅ Features like diarization, summarization, language/topic detection, and more. We’ve raised over $86 million and have been recognized as an Inc. Best Workplace (2021 and 2022), a Forbes Top 50 AI Company to Watch (2021), and a CB Insights Top 100 AI Startup (2021), among others. Come join our revolution to unlock the power of voice technology for everyone. We want to hear what you’ve got to say! An NVIDIA partner and Y Combinator company

Voicera is building the data layer for trust analytics, empowering People to score sincerity in human communication objectively at scale. Why now? Trust is the new currency. • Problem: All video meetings are recorded and processed by LLMs. How people communicate is lost in transcription, causing misinterpretation, and bad decisions. • Solution : Sincerity™ by Voicera analyzes people's behavior to assess how truthful they are from non-verbal cues with 80% accuracy. • Use Cases: Sales (customer insights, revenue forecasts), Training, HR (recruitment), Internal Fraud Investigations… • Outcomes: Voicera empowers customers to make more informed, ethical decisions in situations where trust is critical.

At 4Catalyzer we strive to transform 21st century medicine by solving today’s most challenging problems – across life science research tools, medical devices and therapeutics. We aspire to create significant societal impact and are motivated by the idea that our products will change the lives of people we love. Fundamentally, we are focused on developing new types of sensors and pairing them with deep learning to get a window into biology and medicine that nobody has ever had before. We have audacious plans – but we have an unbeatable team and a founder that has disrupted biomedical science successfully more than once. If you have experience in image reconstruction, mixed-signal IC design, high-speed electronics, nanophotonics, microfabrication, MEMs, biophysics, biochemistry, molecular biology, deep learning, cloud computing, embedded systems, or hold any dual degree with distinction along with computer science please contact us and “put a dent in the universe”.

AssemblyAI is a remote-first company building powerful AI models for developers, startups, and enterprises to transcribe and understand their audio data. Our mission is to make production-ready, state-of-the-art AI models accessible to developers and product teams at extremely large scale through simple APIs. Customers, including Spotify, Algolia, Dow Jones, Happy Scribe, BBC, The Wall Street Journal, and NBCUniversal, use our APIs to transcribe and understand millions of videos, podcasts, phone calls, and zoom meetings every day. Our APIs power innovative products like conversational intelligence platforms, meeting summarizers, content moderation, and automatic closed captioning. We're growing at breakneck speed, and recently announced our Series B round. We've raised $63M in total funding, backed by leading investors including Insight Partners, Accel, Y Combinator, Patrick and John Collision (Founders of Stripe), Nat Friedman (Former CEO of GitHub), and Daniel Gross (Entrepreneur & Investor in companies including GitHub, Uber & SpaceX). Our team is made up of problem solvers, innovators, and top AI researchers from companies like DeepMind, Google Brain, Meta AI, Amazon and Apple. They conduct cutting-edge AI research and develop novel algorithms and techniques to continually push the state-of-the-art in speech recognition and NLP. Join our world-class, remote team and help us build an iconic deep learning company!

BenchSci's vision is to bring medicine to patients 50% faster by 2025. The company empowers scientists with the world’s most advanced biomedical artificial intelligence to run more successful experiments. Backed by F-Prime, Gradient Ventures (Google’s AI fund), and Inovia Capital, BenchSci's proprietary technology accelerates science at 15 top 20 pharmaceutical companies and over 4,300 leading research centers worldwide. BenchSci is a CIX Top 10 Growth company, certified Great Place to Work®, and top-ranked company on Glassdoor.

At Giskard, we are building the first collaborative & open-source Quality Assurance platform for AI models that complies with upcoming AI regulations & standards. Founded in 2021, we are an emerging leader in the new market of AI Quality. Giskard helps AI & Business teams collaborate to evaluate & test AI models. We help organizations increase the efficiency of their AI development workflow, eliminate risks of AI biases and ensure robust, reliable & ethical AI models. Our platform is used by dozens of AI teams across industries, both at enterprise companies & startups. We recently raised our first round of 1.5 million euros, led by Elaia, with participation from Bessemer Venture Partners and notable angel investors. To read more about this fundraising and how it will accelerate our growth, you can read: giskard.ai/knowledge/news-fundraising-2022 We are assembling a team of champions, and we help each other achieve high standards & goals!

Evolution IQ provides a vertical Ai driven claims automation and intelligence platform for the world’s most innovative insurance companies. We’re delivering cognition and reasoning to one of the most data-rich but analytics-poor industries by unlocking their own 'dark data,' which have never before been ingested or used in a systematic tool. Evolution uses cutting edge advances in Natural Language Processing, transfer learning and unstructured data extraction to read and understand insurance claims like a human investigator would, but find anomalies and inconsistencies as only a computer can. The result is a scalable first line of defense against insurance claims fraud.

Our mission is to make AI trustworthy for all enterprises. We’re building a new kind of ML platform to help enterprises accelerate their adoption of AI by building responsible systems with trust, visibility, and insights built-in.