Regular updates on the latest VC-backed AI startups. Follow along to stay informed!
Perfection 42, a Lithuania-based generative AI solution for game development and animation, raised $1.2M in Seed funding. The deal was led by CoInvest Capital and Open Circle Capital, with the Lithuanian Business Angel Network syndicate participating.
Problem to be Solved
Animation for video game development, architecture, design, or art relies on repetitive manual processes and time-consuming rendering of computer-generated imagery (CGI). Perfection 42 uses machine learning techniques to enable artists, designers, and game developers to create 3D content with ease and speed.
How They Use AI
Perfection42 is an abstraction layer over standard AI technologies. The company mentions that some of its applications involve fine-tuning, which suggests that they are probably using open-source models, like stable diffusion. Use cases vary from 3D scene construction to image denoising, each of which depends on a diffusion model (like DALL-E 2 or stable diffusion), but their secret sauce is how they apply the models to accomplish these multi-faceted tasks.
Business Model
Perfection 42 does not publish details on pricing and packaging but likely operates a freemium model similar to Canva which also sells to designers. The company also operates in part as a consulting shop, selling custom dev work, although they are likely to discontinue this as they use the new funding to scale.
Portrait Analytics, a Boston-based personal research assistant for public market investment analysts, raised $3M in Pre-Seed funding. The financing was led by .406 Ventures with participation from partners at leading hedge funds.
Problem to be Solved
Buy-side analysts are chronically time-strapped while consuming data from company filings, transcripts, and research to inform investment decisions. Portrait aims to save analysts time by leveraging machine learning applications to search, summarize, and synthesize investment content.
How They Use AI
Although Portrait Analytics gives us zero hints about their AI stack, it is highly likely that they are using a generative AI chat assistant (e.g. ChatGPT) combined with semantic search through a vector database (maybe Chroma?) to locate the correct context for the question at hand. This is a straightforward implementation with tools like LangChain.
Business Model
Presumably, the company will sell SaaS to investment firms. We’ve requested a demo on the website, but haven’t heard anything yet (understandable as founders David Plon and Connel Gough currently lead a team of just two engineers).
Narrato, a Delaware-based AI content creation & collaboration platform, raised $1M in Pre-Seed funding. The deal was led by AirTree Ventures, with OfBusiness and Shreesha Ramdas participating.
Problem to be Solved
Marketing teams are slowed down by planning, generating, collaborating on, and publishing content. Narrato automates pieces of this workflow by embedding generative AI into a marketing and content creation platform (think Clickup meets Jasper).
How They Use AI
Narrato uses AI to generate, summarize, and rephrase content. They are likely using external APIs from companies like Cohere, AI21, and OpenAI. Their proprietary IP is the set of prompt templates designed to optimize the output of the model on the given task.
Business Model
Classic SaaS. Narrato offers a free plan for individuals, a $9/ month plan for small teams or solopreneurs, and a $19/ month plan for larger businesses.