The Sierra blog

A more reliable inference layer for foundation models

Foundation models are still less reliable than traditional web services, with more downtime and response times measured in seconds, not milliseconds. And as adoption grows, they’re accessed through a burgeoning ecosystem of providers and applications, which have similar functionality but different levels of reliability. Last year, we turned this bug into a feature—developing an adaptive routing client that dynamically selects providers to maximize uptime, minimize latency, and improve the overall experience.

Five ways to use AI agents: healthcare

In the year since Sierra’s launch, we’ve seen how AI can help deliver better, more human care while also reducing costs for the industry.

𝜏²-bench: benchmarking agents in collaborative real-world scenarios

𝜏²-bench challenges AI agents not just to reason and act, but to coordinate, guide, and assist a user in achieving a shared objective. This leap from solo operation to co-ownership of a task pushes agents into a much more demanding space. And, critically, it reflects the kinds of tasks AI agents are increasingly being asked to perform in the real world.

Building more human voice experiences

Here are three lessons we’ve learned building great voice experiences with our customers, for their customers. And what’s so interesting is that it’s many little things like these—not one big thing—that’s helped create a better, more human voice experience.

Serving customer experience and engineering teams—all from one platform

Sierra’s Agent OS is the only platform that supports both no code and programmatic agent development. Our unique approach enables companies to mix and match—choosing the right approach for their business needs, and scaling their agent development as those needs evolve.

Sierra expands to New York

We’re excited to share that Sierra is opening a New York office, and we’re hiring across agent development and sales.

Change agents: Minted

Since we first started working together last year, AI has become central to Minted’s long-term innovation strategy—with information from AI conversations helping to inform their marketing, engineering, and product teams, as well as solving real customer needs with care and creativity.

Lasting relationships: AI agents drive customer retention

By building stronger relationships between companies and consumers, AI agents can help reduce churn—transforming customer service from a cost center into a revenue-generating function.

London calling

We are excited to announce that Suveer Kothari will be joining Sierra as head of our new(ish) London office, which serves companies across the UK and Europe.

Five ways to use AI agents: financial services

Services like banking and insurance play a critical role in our lives, and whatever the issue—whether you’re locked out of your account, disputing a charge, trying to find the right tax information or making a claim—you want to know you’re in the best hands. And that’s where AI agents can really help.

What is an AI agent? And why does my business need one?

Agents are software that can complete tasks on their own, without supervision, just like a human. And AI agents solve the age-old tension between better experiences and higher costs—so you can serve more customers, more often, and in more meaningful ways.

Sierra acquires Receptive AI

We want to ensure that Sierra has the best voice capabilities, and it’s why we’re excited to have acquired Receptive AI. Their voice technology is already integrated into Sierra, enabling companies to build AI agents once and deploy them across any channel.

Same platform, different personalities

Your AI agent is an extension of your brand’s voice, values, and customer experience. As Harmony, Duncan Smuthers, and Barbra all show, Sierra can help your business build better, more human customer experiences with AI.

Sierra’s 𝜏-bench is shaping the development and evaluation of agents

We explore how Sierra’s 𝜏-bench is shaping the development and evaluation of agents—from academic research to industry applications and next-generation development.

Gardening week: evergreen engineering

At Sierra, we value intensity and craftsmanship—Gardening Week is an investment in the long-term quality of our software.

Outcome-based pricing for AI agents

AI agents executing processes autonomously enable an entirely new pricing model, where you pay only when the software achieves specific, valuable outcomes: outcome-based pricing.

Announcing the Sierra APX program

The APX program is an 18-month rotational program that gives new graduates in computer science or a related field the opportunity to work closely with our customers, building AI agents as both an Agent Engineer and an Agent Product Manager.

Sierra speaks

Voice is here. Sierra's AI agents can now engage consumers with rich, natural conversations over the phone—no wait times, no frustration. Learn how this new capability is reshaping the future of customer service.

Bye bye, bots! 5 reasons to say “Hello!” to AI agents

Chatbots continue to fall short of customer and company expectations, struggling to solve problems, communicate with clarity, or self-improve over time. Sierra’s AI agents offer a superior alternative by handling complex tasks, communicating using natural language, and continuously improving to meet evolving customer needs. It’s time to move beyond chatbots and elevate your customer experience with AI agents.

Shipping and scaling AI agents

Sierra agents are refreshing SiriusXM radios, debugging Sonos setup flows, and managing OluKai returns and exchanges. In this post, we share lessons from building AI agents at scale.

Meet the AI agent engineer

Agent engineers at Sierra design, build, and maintain our AI agents. They tackle problems at the intersection of software engineering, frontier AI deployment, and the most pressing challenges of our customers.

The holiday rush has started: How OluKai and Sonos gear up for the season with AI

For most direct-to-consumer and retail brands, the holiday season is the busiest time of year. Not only does this season represent a massive revenue opportunity, but demand surges stress-test systems and staffing. Deploying an AI agent makes it possible for companies to get ahead of the rush, scale up, deliver a better customer experience, and drive sales.

𝜏-Bench: Benchmarking AI agents for the real-world

Sierra’s AI research team is on a mission to advance the frontier of conversational AI agents. In this research paper, we present a new benchmark for evaluating AI agents' performance and reliability in real-world settings, with dynamic user and tool interaction.

The Agent Development Life Cycle

An autonomous AI agent is a new type of software that requires a new approach to development. Sierra has developed a novel, integrated approach to developing agents, from development to testing and release management. We call this the Agent Development Life Cycle.

The Guide to AI Agents

AI agents are autonomous software systems that can reason, make decisions, and pursue goals with creativity and flexibility. They’ll soon be everywhere, helping consumers, employees, and businesses alike.

Meet Sierra, the conversational AI platform for businesses