It feels premature to talk about the “good old days” of search engine optimization. But as anyone who has performed a search recently knows all too well, things aren’t what they used to be. Gone are the days of SEO based around keywords that would deliver a higher ranking. Search engines now supply on-demand, AI-generated answers, pulling from seemingly random links to create their custom-tailored replies. This has left developers scratching their heads. How can they figure out how AI weaves together its responses and game that system?
This is exactly what SerpApi, an Austin, Texas-based company providing a data infrastructure platform, has been trying to address. By offering structured access to public search results, SerpApi is at the forefront of this new frontier of SEO in the generative AI era. And by using their almost endless menu of application programming interfaces, SerpApi offers its clients access to valuable data that can help them make themselves more visible to potential clients worldwide.
The AI Search Era
SerpApi offers API infrastructure that enables developers and enterprises to monitor, train, evaluate, and supply fresh search data on how information appears in AI-generated results. Since its founding in 2017, the company has sought to differentiate itself both in terms of technology, but also philosophically. That means it seeks to provide search engine results in real time as well as structured JavaScript Object Notation (JSON) outputs, with precise location targeting. SerpApi is also big on the First Amendment, and the company is committed to open access as pillars of responsible and fair AI.
“At SerpApi, we deeply cherish the values defined by the First Amendment,” CEO Julien Khaleghy says. “Several of us came to the United States precisely because of the freedoms the First Amendment protects.”
The company is now drawing on those technological and philosophical strengths in the AI Search era. The growth of models like ChatGPT, Gemini, and Claude has created total upheaval for traditional SEO. Brands no longer exist in neatly ranked order, but are hidden away somewhere in the workings of the AI model. SerpApi’s APIs allow users real-time access to the data that informs these AI models and to integrate it directly into their own pipelines and systems.
SerpApi’s structured Search Engine Results Page (SERP) data allows enterprises to continuously obtain updates on how products, policies, or narratives appear across search engines and to correct or supplement their Retrieval-Augmented Generation (RAG) indexes accordingly. The company now offers APIs for Google, Bing, Amazon, Baidu, YouTube and other search engines, most recently rolling out a Bing Copilot API and a Google Travel Explore API. It also improved its performance on Google Trends, released a lightweight Google Images API, and added team management for its developer user base. Its highly skilled team of 33 developers never seems to sleep, introducing updates and tweaks on a constant basis.
This has been quite an engineering feat for the company, as developers creating intelligent agents are using SerpApi to retrieve citations or insert search-verified facts into dialogues. These intelligent agents use SerpApi-generated data as a trustworthy pipeline.
From Paris to Texas
Though SerpApi is very much a team effort, CEO Julien Khaleghy is its conceptual architect. Originally from Paris, France (not Paris, Texas) he relocated to Austin, the state’s capital and a major tech hub in the US, to set up and build the company. According to Khaleghy, the impetus for the company came from his own experiences developing iOS applications some years ago.
“I was working on apps to detect things like dog breeds or types of food with machine learning,” he recalls. “At that time, Apple didn’t have an official AI library for its iPhone and it was difficult to make machine learning work on iPhones because they have limited RAM,” he says.
He overcame this challenge by using Google Images to find images, download them, and then train his computer to learn on them. But scraping Google at that time proved challenging, as did navigating the labyrinthine associated HTML and files. The right API could accomplish just that. So Khaleghy decided to build one himself. “That’s how SerpApi was born,” Khaleghy says.
Now, eight years into the company’s journey, he shows no signs of letting up, but rather is committed to making public search data accessible in a clear and ethical way, even as AI reshapes the information landscape. SerpApi sees its focus on transparency as a strength, as it enables developers to inspect how information is sourced, ranked, and presented by search engines and increasingly, by the AI systems that use those results as training fuel.
The company currently has a variety of use cases. These include straightforward SEO tasks, making sense of Google and other engines’ rankings, why some snippets are treated differently, extraction of the rich snippet; local SEO (which is the same as SEO but for Google Maps); machine learning and AI data for model training; personal background check automation, for use by law enforcement for warrants, for example; company background check automation for banks and financial institutions; news monitoring; as a voice assistant to try to use search engine results for certain requests; and for other big data applications.
And there are more services on the way. In addition to its Walmart Search API and Home Depot Search API, the company rolled out a new OpenTable Reviews API that provides structured access to reviews and ratings data from OpenTable restaurant pages, and made numerous improvements to its long list of existing APIs.
Boasting a response time of just over a second from its platform, one could say that SerpApi is a company that never sleeps.
