Two experienced attorneys who spent years working at a major Seattle law firm and later serving as general counsels at technology companies have launched a new legal practice designed to leverage artificial intelligence for startup clients. Sam Shaddox and Matt Souza observed that traditional law firms were charging emerging companies
substantial fees for services they believed AI technology could perform more efficiently and affordably.

Their newly established Seattle firm, Talairis Law Group, which debuts this week, operates on the premise that artificial intelligence can perform many tasks traditionally handled by junior attorneys at large firms, enabling startups to access quality legal services without the premium pricing associated with established practices.

“We’ve been on the other side of the aisle, and now it’s time for us to make a mark,” Shaddox explained, describing their venture as perfectly timed for the current artificial intelligence landscape and addressing needs within Seattle’s startup ecosystem.

While the concept isn’t entirely novel—venture capital has funded numerous AI-focused legal startups in recent years, including Crosby, Manifest OS, Eudia and Lawhive, which have collectively raised hundreds of millions—Shaddox and Souza identify a critical market gap. According to them, existing competitors have focused on specialized single practice areas such as contract analysis, immigration matters, or merger and acquisition due diligence.

“They’re all picking one lane,” Souza noted, “and there’s not an AI-powered law firm that you can rely on to help you with your day-to-day as things come up, helping to pilot your ship.”

Both founders previously held partner-track positions at Perkins Coie, the well-known Seattle-headquartered law firm. Shaddox subsequently held legal positions at Big Fish Games and OfferUp before becoming general counsel at SeekOut, an AI-driven talent intelligence platform. Souza worked as senior counsel at Zillow before taking the general counsel role at Wrapbook, an entertainment industry payroll and financial services platform.

Their in-house experiences crystallized the problem they now aim to solve. “We were getting billed out the ears for work that—as we were adopting AI in-house—we saw law firms were not doing, or not doing it very well,” Souza said, adding that the fundamental economic structure of traditional law firms is fundamentally flawed.

Talairis operates using what the founders describe as a four-tier system. The foundation consists of a large language model providing the AI capabilities. Above that sits an agentic layer containing over 100 specialized AI agents designed for various legal tasks startups might require. The third layer, termed the “client genome,” maintains a comprehensive profile of each client’s business operations, risk preferences, contractual arrangements and operational history, ensuring customized rather than generic guidance. At the top tier, Shaddox and Souza personally review and approve all deliverables.

“You’re not getting one-off advice that doesn’t know what your company is or does or how it thinks and operates,” Shaddox said. “You’re getting bespoke outcomes.”

The founders cite SAFEs—simple agreements for future equity, commonly used for startup bridge financing—as an illustrative example. First-time founders frequently attempt to manage these independently or engage outside counsel charging $1,500 hourly. Both approaches involve labor-intensive manual work prone to errors when processing notes, side letters and capitalization table calculations.

Talairis has developed a dedicated agent for this purpose. According to the founders, clients submitting a SAFE receive comprehensive deliverables beyond standard legal analysis, including a complete capitalization table incorporating the new note and side letter provisions, with projections showing impact on subsequent financing rounds.

The firm launches with paying clients, though names remain
undisclosed. Currently bootstrapped with just the two founding attorneys, Talairis offers hourly rates approximately half those of typical large firm attorneys. Shaddox states the AI enhancement multiplies productivity sufficiently that effective client costs represent a fraction of traditional legal expenses.

Regarding data privacy concerns—particularly relevant when startups share sensitive legal documentation—Shaddox is unequivocal: “The answer is no. Your data is never used to train a model.” The firm has integrated confidentiality and attorney-client privilege safeguards into its fundamental architecture.

The launch coincides with Anthropic’s release of Claude for Legal, featuring over 20 connectors and 12 practice-area plugins designed for law firms and in-house legal departments. Shaddox views this timing as market validation, noting that while large language models provide a foundational layer, Talairis differentiates itself through proprietary enhancements, elite attorneys, client-specific customization, privilege protections, and agentic architecture that generic plugins lack—elements that transform basic AI tools into optimal legal counsel for startups.