Assembled’s cover photo
Assembled

Assembled

Software Development

San Francisco, California 7,353 followers

The support operations platform | Enabling great support, from workforce planning to AI automation

About us

Industry leaders like Etsy, Robinhood, and Stripe trust Assembled to provide customer-facing AI agents and workforce planning at scale. We automatically resolve millions of interactions through chat, email, and phone while optimizing staffing for hundreds of thousands of support professionals. Our mission is to elevate customer support through AI-powered software that makes life easier for customers and employees.

Industry
Software Development
Company size
51-200 employees
Headquarters
San Francisco, California
Type
Privately Held
Founded
2018

Locations

Employees at Assembled

Updates

  • View organization page for Assembled

    7,353 followers

    This is a special one for us, and for support teams everywhere. Today, we’re launching Assist Omnichannel AI Agent, which brings seamless automation to chat, voice, and email — all in one platform. We’re certain this is the best AI agent for support teams out there. No other AI agent on the market can handle the level of complexity that Assist can, alongside and in place of your human agents. Smarter automation, seamless handoffs, and speedy resolution. You can only find that with Assembled. Read more here: https://lnkd.in/gnNnQjcS

    • No alternative text description for this image
  • Joe Gershenson has spent his career at the forefront of AI innovation — leading ML teams at Stripe and Adept, co-founding a Y Combinator-backed startup, and always chasing the next big shift in technology. His focus? Building AI that solves real problems for real people. That’s what led him to Assembled. After years of seeing AI’s potential to reshape industries, Joe saw something different in Assembled’s approach to AI-powered customer support. Instead of replacing human work, we’re making it more meaningful — helping support teams resolve real issues, not just automate them away. “It has consistently been my experience throughout my career that it's easier to add a flashy technology hype layer on top of a rock-solid business that deeply understands the customer problem than to do the reverse,” Joe says. Now, as our Head of Engineering, Joe is leading the charge. Read more about his journey and why he chose Assembled.

    Engineering the future of AI-powered support: A conversation with Joe Gershenson

    Engineering the future of AI-powered support: A conversation with Joe Gershenson

    assembled.com

  • AI is everywhere in support right now — but how do you cut through the hype and make it work for your team? From inflated promises to unclear ROI, evaluating AI in support can feel like a maze. That’s why we’re breaking it down in our next webinar: The buyer’s guide to AI in support: Building the case, proving the value. Join Assembled AI product expert Brian Yeh as he covers: ✅ How to build a strong AI foundation (before automating anything) ✅ What to look for in AI pricing (and the hidden costs to avoid) ✅ How to structure a proof of concept that actually proves value ✅ The must-haves for your RFP (red flags included) Plus, you’ll get a live demo of Assembled Assist in action — see how top support teams are automating up to 50% of cases and reducing escalations by 20%. Can’t make it live? No problem — all registrants get the recording. But if you’ve got burning AI questions, this is your chance to ask. Register now: https://lnkd.in/gfaBQ23x - Q1 - Webinar_AI_Buyer's_Guide&utm_source=linkedin&utm_medium=paidsocial&utm_content=buyers-guide-webinar

    • No alternative text description for this image
  • Embarking on an AI proof of concept (PoC) is a strategic move, but its success hinges on careful planning. To ensure meaningful results, consider these steps: 👉 Define clear objectives: Identify specific support areas where AI can make a tangible impact, such as automating responses or assisting agents. 👉 Select relevant use cases: Focus on scenarios that align with your customer experience goals, whether it's handling high-volume inquiries or providing agent support. 👉 Establish success metrics: Determine how you'll measure outcomes — look at metrics like first response time, automation rates, or resolution efficiency. By thoughtfully structuring your PoC, you can avoid the "garbage in, garbage out" pitfall and set the stage for AI to enhance your support operations. For a comprehensive guide on executing a successful AI PoC, read our full article.

    How to think about the “proof” of your AI proof of concept (POC)

    How to think about the “proof” of your AI proof of concept (POC)

    assembled.com

  • AI support tools don’t follow a standard pricing model — some charge per interaction, others by resolution, and some bundle features in ways that aren’t always transparent. That makes it easy to end up paying more than you expected or for functionality you don’t actually need. Before signing on the dotted line, it’s worth digging into the details: 👉 What exactly are you being charged for — users, interactions, resolutions, or something else? 👉 How does pricing scale as your support volume grows? 👉 Are there hidden costs, like fees for additional training data or integrations? AI can be a game-changer for support teams, but only if the pricing model works for you. Otherwise, you risk unexpected costs that don’t align with the value you’re getting. Here’s how to ask the right questions before committing.

    How (most) AI support tools are priced, and the questions you need to ask before buying

    How (most) AI support tools are priced, and the questions you need to ask before buying

    assembled.com

  • Last chance to register! Tomorrow, Olivia Teich, Assembled’s Head of Product, is breaking down what’s next for support teams in 2025 — and you don’t want to miss it. We’ll be diving into: 🔹 How to scale support without losing quality 🤖 Where AI actually fits into the customer experience 📊 Smarter workforce management strategies for hybrid teams If you lead a support team (or just want to stay ahead of the curve), this one’s for you. Sign up now: https://lnkd.in/gynUMnT6

    • No alternative text description for this image
  • What happens when you put 250+ ambitious founders, builders, and investors in a room? You get the biggest Founders You Should Know event yet — and an incredible conversation about the future of AI. Assembled Co-founder and CTO John Wang took the stage this week to share how we’re building the AI-powered operating system for support teams, alongside some of the most exciting voices in AI today. If you're an engineer excited to tackle real-world AI challenges, we want to meet you. Check out our open roles: https://lnkd.in/gMRzFVDC 

    • No alternative text description for this image
    • No alternative text description for this image
  • Knowledge bases aren’t just for human agents anymore — they’re now the foundation for AI-powered support, too. Most support teams have a KB that’s grown over time — outdated articles, inconsistent formatting, and gaps in documentation can make it hard for both humans and AI to find the right information. AI thrives on structure, which means a well-organized, regularly updated knowledge base makes a world of difference. So, how do you optimize your KB for AI? Start with these questions: 👉 Are your articles clearly written and formatted? 👉 Is your information structured in a way AI can parse? 👉 Do you have a regular maintenance process? If AI-generated responses aren’t quite hitting the mark, your KB might need a tune-up. Here’s how to fix it.

    A guide to optimizing your knowledge base for AI

    A guide to optimizing your knowledge base for AI

    assembled.com

  • What’s next for support ops in 2025? Let’s talk about it. Customer expectations are evolving. AI is changing the game. And hybrid teams — mixing in-house, outsourced, and AI agents — are the new normal. So how do support leaders keep up without overloading their teams? Join Assembled’s Head of Product, Olivia Teich, to unpack insights from our State of Support Ops in 2025 report. In this webinar, we’ll cover: ✅ How to scale support without sacrificing quality ✅ The evolving role of AI in customer service ✅ Workforce management strategies for 2025 and beyond Can’t make it live? No worries — all registrants get the report and a recording. But we’d love to have you join the conversation! Register now: https://lnkd.in/gJ23WMhM

    • No alternative text description for this image
  • We get a lot of questions like, "What should our adherence be?" and, "How are teams like mine doing when it comes to SLA?" Well, today we have answers. Using aggregated and anonymized performance data from our customers, we created a report with industry-specific benchmarks you can use to uncover inefficiencies, highlight wins, and build a stronger case for the resources your team needs. Ready to see how your team stacks up? Download the report to find out!

    2025 customer support benchmarks

    2025 customer support benchmarks

    assembled.com

Similar pages

Browse jobs

Funding