Key takeaways
- Modem raised $4.4M led by Accel (Feb 2026); as of June 2026 no Series A — repositioned from 'auto-triage PM' to 'product execution for the AI era'
- Founded by Ben Vinegar (ex-Sentry VPE) and Mike Clarke (ex-Treasury Prime VP Eng) — deep devtools DNA, team now ~7
- Pricing went public: free Dial-Up tier, DSL $144/mo, Cable $560/mo, Enterprise custom — event-based usage with unlimited users on paid tiers
- Shipped fast post-funding: Agent Skills, Event Automations, Automation Templates, plus Intercom, PostHog, and Notion integrations
- Integration surface grew to 20+ including Zendesk, Jira, Cursor, and Devin, with Claude Code support announced as coming soon
FAQ
What is an auto-triage product manager?
An AI system that automatically categorizes and prioritizes product feedback from channels like Slack, Discord, and GitHub — eliminating manual triage work and helping teams act faster on user needs.
Why is product management a bottleneck for AI-native teams?
AI coding tools accelerated code production, but feedback collection, backlog curation, and customer communication remain manual. Product management is now the constraint on shipping velocity.
How do AI product tools close the loop with customers?
They automatically notify affected customers when requests ship, generate tailored release notes, and send personalized digests — critical when deploying faster than humans can manually communicate.
What signals matter for prioritizing product feedback?
Customer context (company size, tier, revenue), request frequency across users, emerging trends before they become fires, and correlation with strategic goals.
How much does Modem cost?
Modem offers a free Dial-Up tier (2,500 events/month, 3 users), DSL at $144/month (25,000 events), Cable at $560/month (100,000 events), and custom Enterprise pricing. Paid tiers include unlimited users with a 14-day free trial.
Executive Summary
Updated June 11, 2026.
Modem is positioning itself as the product management layer for teams shipping with AI coding agents. As of June 2026, the company has broadened its tagline from "auto-triage PM" to "product execution for the AI era" — an AI teammate that knows what customers want, what the team is building, and how it's landing (modem.dev).
Founded by Ben Vinegar (founding engineer and VPE at Sentry) and Mike Clarke (VP Engineering at Treasury Prime), Modem raised $4.4M led by Accel in February 2026 (announcement). No Series A has been announced as of June 2026. Backers include founders from Sentry, Framer, Vercel, and Cohere — a who's who of modern devtools.
Since the raise, Modem has shipped at a fast clip: public pricing (free to $560/mo), Agent Skills, Event Automations, Automation Templates, and integrations with Intercom, PostHog, and Notion (blog).
Key Insight: Coding isn't the slow part anymore. The product process is.
Company Overview
| Attribute | Details |
|---|---|
| Founded | 2025 (estimated) |
| HQ | Not disclosed |
| Funding | $4.4M Seed (Feb 2026); no Series A as of June 2026 |
| Lead Investor | Accel |
| Team Size | ~7 (based on About page, June 2026) |
| Stage | Early ("dozens of organizations" per Feb 2026 announcement; no newer public count) |
| Compliance | SOC 2 certified (per About page) |
Leadership
Ben Vinegar — Co-founder, CEO
- Founding Engineer at Sentry
- VP Engineering at Sentry
- Deep experience scaling devtools
Mike Clarke — Co-founder, CTO
- VP Engineering at Treasury Prime
- Background in fintech infrastructure
Notable Investors
Institutional:
- Accel (lead)
- Inovia Capital
Angels (devtools heavyweights):
- David Cramer, Chris Jennings — Sentry
- Ivan Zhang — Cohere
- Koen Bok, Jorn van Dijk — Framer
- Lindsey Simon — Vercel
- Armin Ronacher — Flask, Earendil
- Jacob Thornton — Bootstrap, Pierre Computer Co
- John Laban, Kenneth Rose — OpsLevel
- Dave Johnson — PhoneGap/Cordova
The investor list reads like a who's who of devtools. This is a team with deep credibility in the developer tooling space.
Product Analysis
Core Capabilities
1. Chaos into Clarity
Modem connects to communication channels and builds a "context graph" from product discussions:
- Auto-clustering: Bug reports and feature requests grouped without manual tagging
- Auto-prioritization: Topics ranked by importance
- User/Company Profiles: Understand who matters and who can wait
- Trend Detection: Spot emerging issues before they become fires
2. Act Fast with Ease
Conversational AI agent that teams can talk to directly:
- Create context-rich tickets from Slack/Discord
- Delegate tasks to engineers or coding agents
- Answer complex questions about user needs in seconds
- Grounded in months of user context (not just a sparse ticket description)
3. Close the Loop
Critical for teams deploying faster than they can communicate:
- Notify customers when their requests ship
- Generate release notes tailored for each stakeholder
- Personalized updates and digests for team alignment
4. Agent Platform (new since February 2026)
Modem has layered an automation platform on top of the core triage product (blog):
- Agent Skills (May 2026): teach the agent consistent workflow handling across chats and automations
- Event Automations (March 2026): agent actions triggered by product and GitHub events
- Automation Templates (May 2026): pre-built automations installable in a few clicks, updated by Modem
- Notion-powered workflows (June 2026): agent can create pages and update databases directly in Notion
Integrations
The integration surface has grown from 5 channels at launch to 20+ as of June 2026 (modem.dev):
| Category | Integrations |
|---|---|
| Chat/Social | Slack, Discord, Email; X (coming soon) |
| Customer Support | Zendesk, Intercom, Jira Service Desk |
| Project Management | Linear, Jira, Notion, GitHub |
| Analytics | PostHog; Amplitude, Mixpanel (coming soon) |
| Coding Agents | Cursor, Devin; Claude Code (coming soon) |
| Other | Web search, image generation, MCP support |
Technical Architecture
Modem describes their system as a "real-time product graph":
- No manual tagging: Self-organizing based on conversation content
- Context compounding: Every conversation makes the graph smarter
- Connective layer: Bridges gaps between existing tools
Pricing (public as of June 2026)
Pricing was undisclosed at launch; it is now published (pricing page):
| Tier | Price | Events/mo | Users | Retention |
|---|---|---|---|---|
| Dial-Up | Free | 2,500 | Up to 3 | 30 days |
| DSL | $144/mo | 25,000 | Unlimited | 180 days |
| Cable | $560/mo | 100,000 | Unlimited | 365 days |
| Enterprise | Custom | Custom | Unlimited | Custom |
All paid tiers include a 14-day free trial (no credit card) and pay-as-you-go overages with customer-set monthly budgets. Pricing is event-based rather than seat-based — a notable choice that aligns cost with feedback volume, not headcount.
Market Position
Target Customer
Dev teams that:
- Ship software with AI coding tools
- Have product feedback scattered across multiple channels
- Lack dedicated PM capacity or want to augment PMs
- Need to close the loop with customers faster
Stated Customers
Logos on homepage as of June 2026: Superwhisper, Kernel, Sentry, Zumper, VLT, Basedash, Mycroft (modem.dev)
Since February 2026, DevCycle and Laminar have rotated off the homepage; Zumper and Basedash were added. (Note: Sentry presence is notable given founder's background — likely a friendly/validation customer.)
Competitive Landscape
| Category | Competitors |
|---|---|
| Traditional PM Tools | Productboard, Aha!, Canny |
| Feedback Aggregation | Pylon, Plain, Intercom |
| AI PM Assistants | Various early-stage |
Modem's differentiation:
- Auto-triage — No manual categorization required
- Context graph — Compound learning from conversations
- Full lifecycle — Input (feedback) → Output (release notes/notifications)
- Devtools DNA — Built by people who lived the problem at scale
Strategic Analysis
Strengths
- Founder credibility: Sentry alumni with proven track record
- Investor quality: Best-in-class devtools angels
- Timing: AI coding tools are creating a product management bottleneck
- Full-stack approach: Captures feedback AND closes the loop
Cautions
- Early stage: "Dozens of organizations" (Feb 2026) remains the only public traction figure; no updated customer count, case studies, or revenue signals as of June 2026
- Crowded adjacent space: Feedback tools, PM tools, and AI assistants all converging
- Event-based pricing opacity: What counts as an "event" determines real cost; teams with high-volume Slack/Discord channels could hit limits quickly
- No independent reviews: No substantive community discussion on Hacker News, Reddit, or Product Hunt found as of June 2026 — all public narrative is company-authored
What Developers Say
No substantive third-party community discussion of Modem was found on Hacker News, Reddit, X, or Product Hunt as of June 2026 — searches surface only the company's own funding post and derivative coverage. For a devtools-pedigree team four months post-announcement, the absence of organic developer chatter is itself a data point worth watching.
Open Questions
- How does the auto-clustering perform at scale with noisy channels?
- What's the integration depth with Linear/GitHub (bidirectional sync)?
- How do they handle conflicting feedback from different customer tiers?
- What counts as an "event" in practice, and how does cost scale for chat-heavy teams?
- Will the Claude Code / Amplitude / Mixpanel "coming soon" integrations ship?
Bottom Line
Modem is betting that the product management layer is the next bottleneck as coding accelerates. Four months after the seed announcement, execution velocity is the standout signal: public pricing, an agent skills/automation platform, and a 4x larger integration surface — all shipped by a team of roughly seven (blog, about).
Recommended for: Small-to-mid dev teams with feedback scattered across Slack/Discord/support tools who want auto-triage without dedicated PM headcount — the free tier and $144/mo DSL plan make trialing low-risk.
Not recommended for: Teams needing a proven, referenceable platform — there are still no public case studies, independent reviews, or traction figures beyond February's "dozens of organizations." Large enterprises should wait for the Enterprise tier to mature.
Outlook: The repositioning from "auto-triage PM" to "product execution for the AI era" widens the ambition from triage tool to agent platform. The risk is the quiet demand side: no Series A, no updated customer count, and no organic community discussion yet. Watch for customer case studies with measurable impact, shipped coding-agent integrations (Cursor/Devin/Claude Code), and any follow-on funding.
Research by Ry Walker Research • methodology
Disclosure: Author is CEO of Tembo. Modem operates in an adjacent but distinct space (product management vs. agent orchestration).