What Is Duckie AI? Features, Pricing and Alternatives (2026)

Duckie (duckie.ai) is a four-person Y Combinator startup from the Winter 2024 batch that sells "the agentic platform for support teams": support staff write AI agents in plain English that resolve tickets, investigate issues and pull context from Slack, Jira and Confluence. Duckie publishes no pricing; its /pricing URL returns a 404 and every route leads to booking a call.

We build AI agents for customer service ourselves, on top of the help desk a team already runs. We checked duckie.ai, its customer page and its YC profile on 24 September 2026; the capability claims below come from Duckie's own pages and its case-study numbers are vendor-reported.

What is Duckie?

Duckie's homepage: "the agentic platform for support teams."

Duckie was founded in 2023 by Valerie Li and Joel Ritossa and went through Y Combinator's Winter 2024 batch. Li was a tech lead at Netflix on the password-sharing project that, per her YC profile, "added 8m new subscribers and $200m in revenue"; she also worked at LinkedIn. Ritossa was an AI software engineer at C3 AI. Third-party trackers put Duckie's seed funding at $500,000 as of April 2024, a figure Duckie does not state on its own site. Its YC profile lists a team of 4.

The homepage pitch has moved from "AI support agents" toward a platform: support professionals become "AI builders, shipping agents that resolve tickets, investigate issues, and automate any workflow behind support." The YC profile still describes the original angle, connecting to Slack, Jira and Confluence to generate issue resolutions and write back to documentation. That is a specific bet: most support AI vendors are built for FAQ-shaped consumer questions, while Duckie targets tickets that need context from an engineering team's internal tools.

Duckie's customer page claims a 90% ticket automation rate at Grid, a 95% resolution rate at Vanquish Trader, a 92% ticket deflection rate at Automox, a 66% support cost reduction at Abbiamo and "10x faster workflow setup vs. Zendesk AI" at Medtelligent. These are vendor-reported, customer-specific figures, not an audited average; treat any single number as what one customer achieved.

Duckie's customer stories page: real named logos backing the case-study figures.

How do Duckie's AI agents work?

Duckie's agents are defined in plain English instead of a flowchart builder, and the homepage says no engineers are needed to configure them. Underneath that, a few named mechanisms do the work:

• Multi-model support: "every frontier model, switch whenever you want," so no single model is locked in.
• Tool connectivity via MCP, APIs and native connectors, naming Zendesk, Stripe, Slack and Intercom, plus an MCP gateway.
• A policy engine and PII redaction: policies with permission scoping apply to every agent, and PII is redacted before any tool call.
• A sandbox and a kill switch: agents can't reach customers until proven safe, and any agent can be halted at once.
• Testing, shadow and live modes, plus a full trace of what each agent did.

The staged rollout (testing, then shadow, then live) deserves a closer look. It answers the most common fear about giving an AI agent write access to a support system: that it does something wrong before anyone notices. Shadow mode shows what an agent would have done on live traffic before it's allowed to act.

What are Duckie's key features?

• Plain-English agent definition, with no flowchart or scripting layer.
• Switchable frontier models, with no lock-in to one AI provider.
• Native connectors and MCP/API access to Zendesk, Stripe, Slack, Intercom and similar tools.
• Policy enforcement and PII redaction built into the platform.
• Sandboxed execution with a kill switch for contained agent actions.
• Testing, shadow and live modes for staged rollout on real ticket traffic, with a full trace per run.

How much does Duckie cost?

Duckie does not publish pricing. duckie.ai/pricing returns a 404 (checked 24 September 2026), and the homepage offers "View demo," which leads to a book-a-call page, and a login for existing customers. There is no self-serve signup or rate card.

That fits how Duckie sells: an engineering-facing product for teams that need connectors, policy configuration and a staged rollout, priced through a scoping conversation. A sales-led motion lets the vendor price to what a scoped deployment is worth, at the cost of buyers being unable to compare numbers in five minutes the way they can with vendors that publish prices. We found no third-party contract-value estimate reliable enough to cite. Budget for a sales conversation and a proof of concept on your own ticket queue before you see a number.

What are Duckie's pros and cons?

Pros

• Built for technical support, pulling context from Slack, Jira and Confluence beyond a public help center.
• The staged testing, shadow and live rollout is a useful answer to "what if it does something wrong."
• Multi-model support avoids lock-in to a single AI provider.
• Named connectors to Zendesk, Stripe, Slack and Intercom cover a real technical support stack, well beyond a chat widget.

Cons

• No published pricing; every evaluation starts with a call.
• A small, early-stage team (4 people per its YC profile), worth weighing if you want a vendor with a deep bench for support and reliability.
• Case-study results are vendor-selected and vendor-reported; there's no independently audited average.

Who is Duckie best for?

Duckie suits teams doing technical or DevOps-adjacent support whose tickets need context from Slack threads, Jira issues and internal documentation rather than a public FAQ. If your queue looks more like triaging bug reports and infrastructure questions than answering "where is my order," Duckie's focus on engineering knowledge sources is the thing to evaluate.

It's a weaker fit if your questions are simple and FAQ-shaped, or if you want to see a price before booking a call. Expect a sales-led cycle rather than a self-serve trial.

How does Duckie compare with the alternatives?

Duckie and Decagon both build standalone AI agents that sit apart from a help desk's own ticket queue, though Duckie's angle (engineering knowledge sources, staged rollout) targets a narrower, more technical buyer than Decagon's enterprise-wide pitch. Macha is a different shape again: an agent layer that lives inside Zendesk, Freshdesk, Gorgias, Front, HubSpot or Intercom, working the tickets already sitting there. It's also done for you: the Macha team analyzes your past tickets, builds the knowledge base and the agent, and runs it in safe mode, with drafts as internal notes in your help desk, until the drafts are right, at about $0.40 a ticket, which includes setup and a dedicated success manager who handles your changes. If your queue is technical and your team wants an engineering-facing tool with staged rollout controls, Duckie is worth the sales conversation. Thena is the other vendor chasing that same technical B2B queue. If the tickets that concern you already sit in a help desk and the question is getting AI to work them well, see Macha's pricing for a published number instead of a demo request.

How we researched this

This guide is built from Duckie's homepage, its customers page and its Y Combinator profile, re-read on 24 September 2026. Duckie does not publish a rate card, so the pricing section describes the sales model rather than quoting a price, and we did not obtain a quote. Duckie is demo-gated, so nothing here comes from hands-on use, and G2 turned away our requests.

FAQ

What is Duckie? Duckie is a Y Combinator-backed (W24) startup, founded in 2023 by Valerie Li and Joel Ritossa, that sells an agentic platform for support teams, with agents that connect to Slack, Jira and Confluence to resolve issues beyond what a public help center covers.

How much does Duckie cost? Duckie does not publish pricing. Its pricing URL returns a 404, and every evaluation path leads to booking a call.

Does Duckie offer a free trial? No published self-serve trial exists. Evaluation runs through a demo and, typically, a proof of concept.

What are Duckie's testing, shadow and live modes? They're staged deployment options: test an agent privately, run it in shadow mode on real tickets to see what it would have done without responding, then promote it to live once it's trustworthy.

Which companies use Duckie? Duckie's customer page names Grid, Vanquish Trader, Automox, Abbiamo and Medtelligent in case studies, and its homepage shows logos including Vellum, Wing Bank, Snowplow, Freshpaint, Vapi, Mudflap, Collate, Dynamic and AvantStay.

What results does Duckie report? Vendor-reported, customer-specific figures include a 90% ticket automation rate at Grid, a 95% resolution rate at Vanquish Trader, a 92% deflection rate at Automox, a 66% support cost reduction at Abbiamo, and 10x faster workflow setup than Zendesk AI at Medtelligent. Treat these as individual case studies, not an audited average.

What tools does Duckie connect to? Named connectors and MCP/API access cover Zendesk, Stripe, Slack and Intercom, alongside the Slack, Jira and Confluence knowledge sources its YC profile describes.

What are good alternatives to Duckie? Decagon is a common enterprise-scale comparison for standalone AI support agents, and Thena targets the same technical B2B queue. Parahelp, from Y Combinator's Summer 2024 batch, also sells to software companies and publishes its rates. For a team that wants an AI agent working inside the help desk it already runs, Macha is a different shape of tool worth comparing, and the Macha team sets it up and runs it for you at about $0.40 a ticket, with a dedicated success manager included.

Sources: Duckie homepage, Duckie customers, Duckie on Y Combinator.