The phrase “Agentic AI” is currently at the forefront of discussions. Both large and emerging developers are rapidly creating applications that facilitate the integration of generative AI within specific industries, sparking significant investment interest in these emerging technologies.
A noteworthy example is the Israeli startup, Perfect, a platform designed to aid recruiters in enhancing their candidate sourcing and hiring processes. The company has successfully secured $23 million in seed funding.
Recruiting teams utilize Perfect as a co-pilot, aiding them in crafting job postings, identifying optimal platforms for their listings, and managing incoming applications. The platform both collaborates with and rivals established tools from organizations like Indeed, Recruiter, and LinkedIn.
Perfect claims that its services can save recruiters up to 25 hours of work each week. Since its discreet launch a year ago, the company has expanded its client base from 20 to 200 businesses, including notable names such as Fiverr, eToro, McCann, and Coralogix.
From Facial Recognition to Candidate Selection
Perfect was founded by Eylon Etshtein, who previously led the controversial facial recognition startup Anyvision (which later pivoted, rebranded, and was recently acquired).
According to Etshtein, the inspiration for Perfect stemmed from his experiences at Anyvision, where he took an active role in hiring. He quickly realized that the traditional hiring process was not scalable.
Having founded an AI-driven facial recognition company focused on identifying the elusive “needle in a haystack,” Etshtein envisioned a platform capable of understanding the preferences of Anyvision, ultimately aiding in the hiring process.
After stepping away from his daily responsibilities at Anyvision during a turbulent period—prior to the surge of interest in “resilience” technology for government and military—he recognized the direction he wanted to pursue next.
The AI-driven HR startup space is crowded, yet Etshtein and his investors are convinced that Perfect stands apart. Unlike many competitors, Perfect developed its platform from scratch without relying on third-party large language models, instead creating its own vector data set and training it with information gathered from external sources. Etshtein mentioned that they frequently purchase data from larger recruitment firms, subsequently “cleaning” it for reuse.
“At the inception of Perfect, ChatGPT had not yet emerged,” he noted. “There was no infrastructure available to construct a career trajectory algorithm that could comprehend your past, analyze your present, and predict your future,” he said.
Despite the lengthy development time of approximately three years in stealth mode to launch the Perfect platform, Etshtein asserts that their foundational work would remain relevant and not be overshadowed by the emergence of large language models. “LLMs are ineffective with extensive data sets,” he stated. In recruiting terminology, a “payload” refers to the extensive array of data collected about a candidate, organized and annotated for insightful analysis.
“We must utilize proprietary data that we annotate; otherwise, we wouldn’t achieve the precision we currently offer,” he emphasized.
Today marks the first public announcement of their funding, which occurs in two waves. A year ago, Perfect secured an equity investment of approximately $12 million from firms including Target Global, RTP Global, and Pitango. Recently, they acquired an interest-free SAFE note, which will convert to equity in the next funding round, from Hanaco Ventures, Joule Ventures, and Young Sohn, the former president of Samsung who sits on Arm’s board.
“In a sector eager for genuine innovation, where both recruitment agencies and candidates are victims of archaic manual processes or subpar AI tools, Perfect is leveraging proprietary data sets and operationalizing industry-specific workflows to revolutionize recruitment practices, automating much of the daily tasks faced by their clients,” stated Lior Prosor, a partner at Hanaco Ventures.
Indeed, recruitment has become a focal point for developers creating AI applications, particularly considering its significant inefficiencies.
Certain job postings or prestigious companies often receive an overwhelming surge of applications, and sorting through the pertinent candidates is—by nature— akin to “finding a needle in a haystack,” as articulated by Perfect’s CEO Eylon Etshtein in a recent discussion.
Conversely, some recruiters struggle to attract applicants, often due to a mix of visibility issues or unpopularity of the job or organization. With a significant workforce dedicated to evaluating applications, it becomes clear why AI developers are focusing their efforts on enhancing recruitment processes.
Perfect is not alone in this field, as competitors include LinkedIn (which offers various AI tools for recruiters and job seekers), along with emerging companies like HiBob, Workable, Maki, Mercor (which recently achieved a $2 billion valuation), Tezi, and SeekOut (which downsized last year) among many others.
Looking ahead, Perfect aims to enhance its toolset for recruiters and also plans to develop a free resource for job seekers, enabling them to refine their search efforts and, in turn, providing the startup with additional data for future initiatives.
Compiled by Techarena.au.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence
