Clay Bavor | $1B+
Get in touch with Clay Bavor | Clay Bavor, cofounder of Sierra, spent 18 years at Google before launching the AI customer-service startup with Bret Taylor. At Google, he most recently led Google Labs and previously helped lead major product efforts including AR/VR, Project Starline, Google Lens, and Google Workspace design and product teams. At Sierra, he brings a product-heavy background to building AI agents for enterprise customer experience.
Clay is Co-Founder of Sierra. Previously, Clay spent 18 years at Google, where he most recently led Google Labs. Earlier, he started and led Google’s AR/VR effort, Project Starline, and Google Lens. Before that, Clay led the product and design teams for Google Workspace.
Sierra is an American artificial intelligence company specializing in autonomous conversational AI agents for enterprise customer service. Co-founded in 2023 by Bret Taylor (former co-CEO of Salesforce and current chair of OpenAI) and Clay Bavor (former Google executive), the company builds customizable AI agents that handle complex, multi-step customer interactions across channels including chat, voice, email, SMS, WhatsApp, and more.[1][2] Sierra emphasizes enterprise-grade solutions with strong security, compliance, and outcomes-based pricing, where businesses pay only for successfully resolved tasks without human intervention.[1][3]The platform includes tools such as Agent Studio for low-code agent building, Agent SDK for advanced development, integrated guardrails, tool integrations, observability via Experience Manager, and features like Agent Memory for personalization using conversation history and customer data.[3] These agents support proactive engagement, multilingual operation in over 34 languages, and high-resolution rates (often 70-90% in deployments), enabling businesses to automate tasks like authentication, financial workflows, returns, and personalized support in regulated sectors such as fintech and healthcare.[4][3]Sierra achieved rapid growth, reaching $100 million in annual recurring revenue (ARR) within seven quarters of its February 2024 public launch and securing a $10 billion valuation following a $350 million funding round led by Greenoaks in September 2025.[5] Earlier rounds included $175 million at a $4.5 billion valuation in October 2024.[6] The company serves major clients including SoFi, Ramp, Rocket Mortgage, Sonos, SiriusXM, ADT, WeightWatchers, Casper, and others, with deployments reaching millions of users and impacting over 95% of Black Friday shoppers, more than 50% of families in healthcare, and significant portions of fintech and media sectors.[5][4]Sierra developed τ-bench (also known as TAU-bench), a benchmark for evaluating AI agents on real-world, dynamic tasks involving user interactions and tool use, to measure performance in complex, collaborative scenarios beyond traditional metrics.[7] The company positions its agents as a step beyond basic chatbots, focusing on deep personalization, business outcomes, and long-term customer relationships rather than short-term cost reduction.[5][1]
Overview
Company description
Sierra is an American artificial intelligence company that develops a conversational AI platform for businesses, enabling them to build autonomous AI agents that deliver better, more human customer experiences.[2][8] The platform specializes in creating agents for enterprise customer service and experiences, allowing companies to deploy sophisticated AI that handles complex, multi-step interactions and performs actions on behalf of customers.[3][8]Sierra's agents operate across multiple channels, including chat, SMS, WhatsApp, email, voice, and ChatGPT, unifying interactions to provide seamless, personalized support.[3] These agents go beyond basic chatbots by reasoning, problem-solving, making decisions, and integrating with business systems such as order management and CRM to execute tasks like subscription upgrades or delivery management.[8] The company emphasizes authentic, brand-aligned, and trustworthy interactions, with built-in safeguards to ensure compliance and security.[8]Positioned as a platform for customer-facing AI agents, Sierra focuses on enterprise-scale deployments that augment human support teams by enabling 24/7 availability, multilingual capabilities, and high-resolution handling of inquiries.[2] In real-world examples, Sierra agents have managed nearly 70% of customer sessions with strong satisfaction scores or handled over half of cases during peak periods, demonstrating their capacity for scalable, empathetic service.[8] Co-founded by Bret Taylor and Clay Bavor, Sierra aims to transform customer experiences across industries through advanced conversational AI.[2]
Leadership
Sierra is led by its co-founders, Bret Taylor and Clay Bavor, who met while working at Google and bring extensive experience from leading technology companies to their roles at Sierra.[2]Bret Taylor serves as CEO of Sierra. He previously served as Co-CEO of Salesforce and currently serves as Chairman of the OpenAI board. Earlier in his career, Taylor co-created Google Maps, served as CTO of Facebook, and founded Quip.[2][1][9]Clay Bavor is co-founder of Sierra. He spent 18 years at Google, where he led Google Labs, initiated and led the AR/VR efforts including Project Starline, and oversaw product and design teams for Google Lens, Google Workspace, and related initiatives.[2][10]The company was co-founded by Taylor and Bavor in 2023.[1]
History
Founding
Sierra was co-founded by Bret Taylor and Clay Bavor in early 2023.[1][11]The company remained in stealth mode until its public announcement on February 13, 2024.[11]Taylor and Bavor first met while working together at Google in 2005.[2][11]The idea for Sierra emerged shortly after Taylor resigned as co-CEO of Salesforce in January 2023 to return to entrepreneurship.[11][1]Bavor, a longtime Google executive who had spent 18 years at the company, was similarly motivated to leave his role amid emerging opportunities in AI.[1][2]During a lunch meeting in early 2023, the two friends discussed their next steps and quickly decided to collaborate on a new venture.[11][1]
Funding rounds
Sierra has raised substantial capital in a series of funding rounds since its public launch in early 2024, reflecting rapid investor enthusiasm for its enterprise-focused AI agent platform.In February 2024, Sierra raised $110 million in an initial funding round led by Sequoia Capital and Benchmark Capital, achieving a valuation of approximately $1 billion.[11][12]In October 2024, the company closed a $175 million funding round led by Greenoaks Capital at a $4.5 billion valuation.[13][14]In September 2025, Sierra announced a $350 million funding round, again led by Greenoaks Capital, valuing the company at $10 billion.[15][16][17]These rounds have brought Sierra's total funding to approximately $635 million.[15][18]
DateRoundAmount RaisedValuationLead Investor(s)
February 2024Initial$110 million~$1 billionSequoia Capital, Benchmark Capital
October 2024Funding$175 million$4.5 billionGreenoaks Capital
September 2025Funding$350 million$10 billionGreenoaks Capital
This funding trajectory has supported Sierra's expansion as a leading provider of autonomous conversational AI agents for enterprise customer service.
Growth milestones
Sierra achieved $100 million in annual recurring revenue (ARR) in November 2025, seven quarters after its public launch in February 2024.[5][19] This milestone marked one of the fastest growth trajectories among enterprise software companies.[5] By early 2026, the company's ARR exceeded $150 million following its first $50 million quarter.[20]Agents built on Sierra's platform achieved substantial scale in consumer reach. Agents touched over 95% of US shoppers, as well as significant portions of other sectors such as 50% of families in healthcare and 70% of the fintech value chain.[20] Sierra-powered agents reached the majority of U.S. consumers across retail and healthcare.[21] More than 50% of the company's customers reported annual revenue exceeding $1 billion, reflecting broad enterprise adoption among large organizations.[20][21]
Platform and technology
Core platform architecture
Sierra's core platform architecture is centered on Agent OS, which employs a constellation-of-models approach integrating more than 15 frontier, open-weight, and proprietary models to power autonomous conversational AI agents capable of complex, multi-step interactions. The system uses modular task abstractions to isolate responsibilities—such as retrieval, classification, tool usage, policy enforcement, and tone management—allowing agents to be composed from specialized, cleanly separated capabilities rather than monolithic structures. Orchestration and routing are handled automatically under the hood, dynamically selecting models optimized for specific task demands like low-latency decision-making, high-precision classification, long-context reasoning, or pitch-perfect conversational tone.[22]This modular design enables agents to execute higher-agency tasks involving advanced reasoning, reflection, and tool use, supported by supervisory layers that enforce guardrails, policies, and quality checks while maintaining reliability through built-in redundancy and automated failover across model providers. As a result, the architecture supports autonomous handling of sophisticated enterprise workflows, including customer authentication, financial transactions, returns processing, and personalized interactions that require context retention and multi-step execution.[22]The platform operates across multiple channels—including chat, voice, email, and SMS—ensuring consistent agent performance and seamless deployment in diverse customer communication environments.[3]
Agent development tools
Sierra provides two main developer-facing tools for building AI agents: Agent Studio, a no-code platform, and the Agent SDK, a programmatic interface for more advanced customization. These tools enable teams to create sophisticated conversational agents tailored to enterprise needs, with Agent Studio targeting non-technical users and the Agent SDK supporting developers who require greater flexibility and control.[23][24]Agent Studio empowers teams to build, configure, and manage agents without writing code, using an intuitive interface that supports natural language inputs and visual configuration. Users define complex workflows through Journeys, composable building blocks that outline agent logic for handling nuanced interactions, such as referencing knowledge sources or calling APIs. The tool includes Simulations for AI-driven testing of agent behavior across scenarios, including voice-specific evaluations like transcription accuracy and noise handling. It also offers Knowledge management to connect agents to external sources like help centers or documents, with features for identifying gaps and ensuring real-time data retrieval. Agent Studio 2.0 introduced Workspaces for collaborative editing with version control and rollback capabilities, as well as an Integration Library that allows non-technical users to connect business systems directly without coding, enabling actions like processing returns or retrieving order data. These features allow rapid iteration and deployment of brand-aligned agents across channels.[23][25][26]The Agent SDK provides developers with programmatic tools to construct agents through declarative definitions, enabling the composition of reusable skills—such as triage, response generation, and confirmation—into multi-step workflows. It supports orchestration for directing agent actions across complex processes, with tuning options to balance creativity and determinism. Developers can integrate real-time knowledge from internal or external systems and perform secure actions like updating records or escalating issues. The SDK includes out-of-the-box AI skills and systems integrations to accelerate development, along with tools for omnichannel deployment, simulations to verify performance, and debugging to inspect logic and API calls. This approach abstracts behaviors from underlying models, facilitating maintenance and scalability while preserving developer control within existing environments.[24][27]Tool integrations form a core part of both platforms, allowing agents to connect with enterprise systems for dynamic data access and action execution. In Agent Studio, the Integration Library simplifies connections for non-developers by supporting credential setup and endpoint configuration. The Agent SDK offers broader extensibility with pre-built integrations and composable tools for handling real-time queries and secure operations, enabling agents to interact reliably with business infrastructure.[24][26]
Security, guardrails, and observability
Sierra emphasizes enterprise-grade security, compliance, and responsible AI practices to enable safe deployment of autonomous agents handling sensitive customer interactions. The platform holds certifications including ISO 27001 for information security management systems and ISO 42001 for AI management systems, alongside compliance with SOC 2, HIPAA, GDPR, CCPA, and CSA STAR standards.[28][29] Customer data remains under strict control, with encryption and automatic masking of personally identifiable information, while agents access systems of record in a deterministic manner to uphold organizational policies.[28] Sierra maintains a Trust Center providing detailed documentation on controls, policies, and compliance practices.[28]Built-in guardrails enable companies to declaratively define and enforce agent objectives, policy boundaries, brand voice, and behavioral constraints. These guardrails restrict agents to approved topics, keywords, and actions, preventing off-limits requests or misuse while allowing deterministic logic for regulated workflows.[30][24] To address non-deterministic LLM behavior, Sierra employs supervisory layers and secondary "supervisor" agents that monitor primary agent outputs for factuality, policy adherence, topic drift, and potential abuse, intervening or escalating as needed.[28][31] This layered approach reduces hallucinations, enhances security, and supports traceability of agent decisions.[29]Observability is provided through the Experience Manager, which allows customers to track agent performance, monitor adherence to instructions, and manage continuous improvement. Interactions are annotated and analyzed to identify issues early, with built-in analytics flagging anomalies and supporting rapid iteration.[32] Agents log detailed reasoning traces for each decision, enabling auditing and oversight, while supervisory monitoring ensures quality and consistency across deployments.[31]
τ-bench benchmark
τ-bench (also stylized as 𝜏-bench or TAU-bench) is a benchmark developed by Sierra's AI research team to evaluate the performance and reliability of AI agents in realistic, real-world settings that involve dynamic interactions with simulated users and domain-specific tools. Introduced in June 2024, it addresses key limitations in prior benchmarks by focusing on multi-turn conversations, policy adherence, and long-horizon task completion rather than single-round or static interactions.[33][34]The benchmark emulates dynamic conversations between a user—simulated by large language models—and a language agent equipped with programmatic API tools and domain-specific policy guidelines. It includes two primary domains: retail (τ-retail) and airline (τ-airline), chosen for their complexity in customer support scenarios that require balancing user requests with strict operational rules. Tasks involve multi-step processes, such as handling reservations, modifications, or inquiries, where agents must interact over extended turns, use tools correctly, and follow policies consistently. The framework incorporates realistic databases, tool APIs, and LLM-generated user behaviors to create diverse and unpredictable interactions.[33][34]Evaluation relies on an objective, stateful method that compares the final database state against an annotated goal state, avoiding subjective measures of conversation quality. A novel metric, pass^k, assesses reliability by measuring the fraction of tasks successfully completed across k independent trials of the same scenario, highlighting consistency issues even among high-performing models. Experiments on leading models, including GPT-4o and Llama 3 variants, revealed that even the strongest agents achieve less than 50% overall success and exhibit significant drops in reliability (e.g., GPT-4o falling to below 25% on pass^8 in the retail domain). These results underscore challenges in rule-following, long-term planning, and handling conflicting information.[33][34]τ-bench has been made publicly available with code, data, and evaluation tools on GitHub, enabling researchers and developers to test and compare agent architectures in realistic conditions. It has influenced subsequent work in agent evaluation by establishing a standard for multi-turn, tool-augmented, and policy-constrained performance.[35]
Customers and applications
Notable customers
Sierra serves a range of prominent enterprise customers across industries including financial services, mortgage lending, consumer electronics, entertainment, home security, and health and wellness.[4]Notable clients include SoFi (personal finance), Ramp (corporate spend management), Rocket Mortgage (home lending), WeightWatchers (health and wellness), Casper (mattress and bedding), Sonos (audio equipment), SiriusXM (satellite radio and audio entertainment), and ADT (home security and automation).[4][36]These and other enterprises have adopted Sierra's platform for customer-facing AI agents, contributing to the company's growth among large-scale organizations.[4]
Use cases and performance
Sierra's AI agents are primarily deployed for autonomous handling of complex, multi-step customer service interactions across chat, voice, email, and SMS channels. These agents excel at tasks requiring reasoning, tool integration, and adherence to business and compliance rules, such as customer authentication, financial transaction processing, returns and refund management, billing dispute resolution, account maintenance, and personalized support.[37][38]In financial services and fintech environments, agents commonly perform identity verification, review suspicious activity for fraud detection, process disputed charges, manage basic account operations (e.g., balance inquiries, password resets, card replacement), and deliver tailored retention offers. These capabilities are particularly valuable in regulated industries, where agents maintain high standards of security, compliance, and accuracy while reducing manual escalation.[37]Real-world deployments demonstrate strong performance, with resolution rates frequently reaching or exceeding 70% for common inquiry types, including product questions, order status checks, and billing issues. Examples include automated resolution rates of approximately 70% on product-related inquiries, over 70% for multi-channel support scenarios, and in some cases significantly higher automation levels, accompanied by customer satisfaction (CSAT) improvements of 20% or more and faster issue handling compared to prior systems.[38][4]These outcomes highlight Sierra's focus on delivering high-resolution, autonomous customer-facing AI that supports enterprise-scale operations in complex and regulated contexts.[37][38]
Industry positioning
Differentiation from chatbots
Sierra positions its technology as a platform for autonomous AI agents that go beyond the limitations of traditional chatbots, which the company describes as inflexible, rule-based systems incapable of handling modern customer service demands.[39]Sierra argues that conventional chatbots frequently fail to understand nuanced customer requests, maintain conversational context, represent brand personality, or adapt over time, leading to frustrating experiences.[39] In contrast, its AI agents are engineered to comprehend complex interactions empathetically, execute multi-step processes autonomously, and evolve continuously to align with business needs and customer expectations.[39]Key differentiators include the agents' ability to manage evolving, related requests within a single conversation—such as investigating issues, proposing tailored solutions, and taking actions like applying discounts or adjusting services—without escalation to human agents.[39] Sierra emphasizes that its agents embody brand tone and personality, deliver personalized and proactive interactions, and improve through ongoing tuning and knowledge integration, capabilities absent in static chatbots.[39][1]Co-founder Bret Taylor has highlighted the goal of creating experiences that are "personal and proactive, but not pushy," with a level of nuance enabled by advanced AI.[1] Clay Bavor has noted the agents' advantage in leveraging context from prior interactions and internal data, starting conversations more effectively than from scratch.[1]Through this approach, Sierra seeks to replace outdated chatbot models with agent-based systems that perform as active, outcome-oriented participants in customer service.[39]
Focus on regulated industries
Sierra has placed particular emphasis on regulated industries, with financial services representing a key focus area due to the sector's stringent requirements for compliance, security, and trust.The company's AI agents support compliance-sensitive tasks in these environments, including fraud detection and dispute resolution, card and account support such as activations and replacements, fee inquiries, and everyday banking services, as well as personalized product recommendations and financial workflows like home refinancing assistance.[40][16]Sierra's platform incorporates security features such as encryption and masking of personally identifiable information (PII) at all times, combined with continuous monitoring, pre-defined goals, guardrails, and built-in quality assurance to enable secure, compliant operations in highly regulated settings.[40][3]This approach allows financial institutions to deliver personalized, efficient customer interactions across channels while upholding regulatory standards and maintaining trust, which the company describes as paramount in the sector.
Disclaimer: This profile is based on publicly available information. No endorsement or affiliation is implied.
Join UHNWI direct Affiliate Program
Earn Passive Income by Sharing Verified Contact Information of Billionaires, Centi-Millionaires, and Multi-Millionaires on the UHNWI Direct Platform
Maximize your earnings potential by sharing direct and validated contact information of the ultra-wealthy, including billionaires, centi-millionaires, and multi-millionaires. Join the UHNWI Direct platform and tap into a lucrative passive income stream by providing valuable data to those seeking high-net-worth connections. Start earning today with UHNWI Direct.
You may also be interested in reviewing other UHNWIs profiles.
To find the person you want to contact, start typing their name or other relevant tags in the search bar.
Please note: Our database contains over 10,000 direct contacts of UHNWIs, and it is highly likely that the individual you are seeking is already included. However, creating individual profiles for each contact is a meticulous and time-intensive process, So, if you are unable to find the profile of the individual you are looking for, please click here.
Filter by Net Worth: All | Billionaires | Centi-Millionaires | Multi-Millionaires
Filter by Location: All | USA | Canada | Europe | UK | Russia & CIS | Asia | MEIA | Australia | Latin America
Filter by Age: 1920-1930 | 1930-1940 | 1940-1950 | 1950-1960 | 1960-1970 | 1970-1980 | 1980-1990 | 1990-2000 | 2000-2010
Related People
UHNWI direct
