The Rise of the AI Engineer

Also: Inflection AI Building 22,000 GPUs 22 exaFLOPS Generative AI Cluster

Author

AI KATANA
July 02, 2023

Hi!

Welcome to today’s issue of AI KATANA. We're excited to share a collection of stories illuminating the transformative role of AI across various sectors. The generational shift to the role of the AI Engineer has emerged as Foundation Models and AI-powered APIs become more accessible. Gen Z workers are capitalizing on AI tools like chatbots to enhance productivity and reshape traditional work patterns. Audi is integrating AI in manufacturing, utilizing an AI system for quality control of spot welds and aligning with its Automotive Initiative 2025. Notably, Inflection AI is constructing the world's largest AI cluster with a breathtaking 22 exaFLOPS performance, thanks to its impressive list of backers. In AI tools updates, we're featuring ByteDance's AI-powered music creation app Ripple and a unique Japanese app that uses AI to detect discomfort in cats. Lastly, in our Venture Capital updates, we shine a spotlight on Voice.ai, a real-time voice changing technology company that has recently raised $6M in funding. Enjoy the read!

Let’s get slicing:

  • 👨🏻‍💻 The Rise of the AI Engineer

  • 👩🏻‍🎤 AI could supercharge the Gen Z takeover

  • 🚘 Audi begins roll-out of AI for quality control of spot welds

  • 🍪 22,000 GPUs: Inflection AI Building 22 exaFLOPS Generative AI Cluster

The rise of the AI Engineer is being observed as a generational shift in applied AI, propelled by the capabilities and widespread availability of Foundation Models. A host of AI tasks that previously required extensive research and development time can now be achieved quickly through APIs. This surge in AI application has led to a multitude of challenges, including model evaluation, tool selection, and staying updated with the rapidly growing body of AI literature. It's becoming increasingly evident that addressing these challenges is a full-time job, leading to the emergence of a new subdiscipline in software engineering: the AI Engineer. This role has been adopted in numerous startups and large companies, with professionals taking AI advancements and turning them into usable products, often in record time. Interestingly, these AI Engineers often come from an engineering background, not a research one. The author predicts that the number of AI Engineer roles will surpass Machine Learning Engineer roles in the near future, changing the perception that AI engineering is a subset of machine learning or data engineering.

Gen Z workers are more frequently using AI tools, like chatbots, to perform tasks such as drafting emails, research, and brainstorming. This generation is seen as well-positioned to take advantage of the AI boom, using these tools to become more productive, efficient, and to focus on more strategic aspects of their work. Their comfort with AI developments and their eagerness to experiment with new tools like ChatGPT could help them outpace older generations. However, while the AI revolution could potentially lead to a reduction in certain jobs, many older workers could still remain competitive if they're open to learning these AI tools. Regardless, experts anticipate Gen Z, having grown up alongside these technological advancements, to adopt these tools more quickly and thus have an edge in the evolving job market.

Audi is advancing its production activities by implementing an artificial intelligence (AI) system for quality control of spot welds in car body construction. This initiative has come after a successful pilot project at the Neckarsulm site. The AI, which analyzes around 1.5 million spot welds on 300 vehicles each shift, allows employees to concentrate on possible anomalies, improving the efficiency of quality control. Preparations are underway to install the technical infrastructure for AI use at other Volkswagen Group plants, including Audi Brussels and the Volkswagen plant in Emden. The data generated by the AI can potentially optimize other processes and be used for predictive maintenance. The development process for this initiative was conducted in collaboration with the German Association for Quality, and the Fraunhofer Institutes for Industrial Engineering and Manufacturing Engineering and Automation.

Inflection AI, a Palo Alto-based startup, has announced that it is constructing the world's largest AI cluster. This cluster, comprised of 22,000 NVIDIA H100 Tensor Core GPUs, will deliver a stunning 22 exaFLOPS performance. The ambitious project is backed by a recent $1.3 billion funding round, which included contributors such as Microsoft, Reid Hoffman, Bill Gates, Eric Schmidt, and NVIDIA, raising the total funding of the company to $1.525 billion. Inflection AI, partnering with CoreWeave and NVIDIA, aims to utilize the cluster for training and deploying large-scale generative AI models. The cluster, which is already being rolled out, is designed to support Inflection AI's Pi - a personal intelligence AI designed for user-friendly interaction and advice. The company, founded in early 2022, is an 'AI Studio' specializing in personal AIs and includes professionals who previously worked at major tech companies like DeepMind, Google, OpenAI, and Meta.

🛠️ AI tools updates

ByteDance, the parent company of TikTok, has introduced an AI-powered music creation app called Ripple, currently available as an invitation-only beta test for iOS users in the U.S. The free app allows users to create, compose, and edit audio, appealing to musicians as a compositional aid and social-media creators for creating background music. Features include a 'Melody to Song' function, which transforms hummed or sung melodies into instrumental accompaniments, and a virtual recording studio for audio editing. The AI model for Ripple has been trained on music licensed or owned by ByteDance, as well as in-house produced music, but not on commercial music from labels or independent artists' tracks from TikTok's SoundOn distribution arm.

A Tokyo-based tech firm, Carelogy, in partnership with Nihon University's College of Bioresource Sciences, has developed an application called "Cat Pain Detector" to identify signs of discomfort in cats. The app, which has gained 43,000 users primarily in Japan, Europe, and South America since its launch last month, was trained using thousands of cat photos. Developers closely analyzed the positions of cats' ears, noses, whiskers, and eyelids, using a scoring system from the University of Montreal to detect differences between healthy cats and those experiencing pain. The AI-driven detection system has refined its skills with approximately 600,000 user-uploaded photos, achieving an accuracy rate of over 90%.

💵 Venture Capital updates

Voice.ai, a company offering real-time voice changing technology, has successfully raised $6 million in funding led by Mucker Capital and M13. The company, which has grown to over 480,000 users and a library of more than 50,000 voice filters via word of mouth and a self-funded $3 million, aims to expand its voice changing tech into new areas. The tools, available on multiple platforms, have been adopted by gamers, content creators, and others, allowing them to create or modify voices for live or recorded use.

🫡 Meme of the day

⭐️ Generative AI image of the day

3