Multimodal search startup Objective Inc. exited stealth mode today, announcing the launch of its platform and the closure of a $13 million venture funding round led by Matrix, with participation from Two Sigma Ventures and others.

Founded by machine learning experts who previously worked at Apple Inc., Amazon.com Inc., Google LLC, Meta Platforms Inc. and Twilio Inc., Objective has created a low-code platform that enables anyone to build “multisearch” applications that work quite differently from traditional keyword-based search, such as Google Search or DuckDuckGo.

One of the first differences is that Objective’s search applications can handle input and output in four different formats, namely text, images, video and audio. More important, though, is how they generate their results. According to Objective, its proprietary search engine can generate contextually relevant results, allowing people to search with their natural voice or combine different images and text queries. What’s more, the system learns over time how to optimize its search results based on the kind of user behavior it sees.

Objective’s technology is based on a technique known as multimodal search, which is a concept that has been around for some years already. It wants to become the go-to solution for all developers looking to integrate multimodal search into their websites and applications.

The startup was co-founded by its Chief Executive Officer Pablo Mendes (pictured, center), who said in a statement that customers today expect search applications to understand their intent in the same way ChatGPT does. “We developed Objective Search so developers can keep up,” he added. “This is what the 1% of the most sophisticated technology companies are working on today, and now it’s available to every website and user experience on the internet.”

Mendes previously worked at Apple’s AI machine learning division alongside his co-founders Joachim Daiber (left) and Lance Hasson (right), where they were tasked with improving Siri’s ability to handle general knowledge queries. “We saw this huge movement happening around machine learning,” Mendes told TechCrunch. “We knew that this was going to change the way that everybody will approach search.”

Objective is launching its main offering with nearly a dozen partners, including the technology news website The Information, officially known as Lessin Media Co., and the creative designer talent platform Dribbble Holdings Ltd.

The Information is one of the most popular and respected technology industry news publications around, and it’s now encouraging its users to search through its website using natural language and complete, meaningful sentences. This functionality is powered by Objective Search, and it’s hoped that it can pave the way to deeper engagement with readers.

The Information’s founder and “intern” Sam Lessin said his company tried numerous different search tools before settling on Objective. “There are multiple approaches – we’ve tried them all, including rolling our own and using GPT4,” he said. “Objective Search is a hassle-free option you can trust.”

Dribbble is a talent showcase platform that’s used by millions of graphics designers and others to share and discover digital artwork. It says Objective’s ability to combine image search and understand metadata that’s often inconsistent or sparse meant that it consistently outperformed its own, internally developed search tools.

“You can only do so much with manual tagging and scrubbing of results,” said Dribbble CEO Zack Onisko. “Implementing semantic search in-house was taking very long. We saw a demo of Objective Search and it was night and day! This partnership will significantly accelerate our data science roadmap.”

Objective’s platform is available now, with pricing based on the size of the customer’s company, how much data they have and the number of search queries they handle each month.

Photo: Objective