Startup Ando came out of stealth on September 24, 2026 with $20 million in seed funding from Accel, Index Ventures, and Emergence Capital, and a straightforward pitch: team messaging built from the ground up so AI agents participate as members with their own identity, not as bots bolted on through an integration.
On platforms like Slack or Microsoft Teams, AI agents typically show up as third-party integrations or bots: they get their own, often limited, slice of permissions and rarely a persistent identity comparable to a human employee's. Ando flips that design: agents take part in channels, threads, and live conversations with their own identity, permissions, and shared context, built so they operate as teammates rather than connected tools.
Ando bills itself as agnostic to whichever agent or "harness" a given team uses: it's compatible with Codex, Claude, Grok, and other agents or frameworks companies already run, rather than requiring every agent to sit on Ando's own proprietary infrastructure. The company says it already has teams using the platform across a dozen countries, in sectors including software, real estate, and financial services, and that its own internal team has worked exclusively inside Ando since January 2026, with agents taking part in engineering, research, product, operations, and sales.
Ando's stated goal is to become the default messaging platform for companies that treat agents as coworkers rather than simply automated tools. It's a bet consistent with a broader 2026 trend: as agents carry out multi-step tasks more autonomously, companies need a place where that work is visible, auditable, and coordinated alongside people's work, instead of happening in silos scattered across different tools.
For a company considering giving its agents a place like this inside its collaboration tools, the governance questions need to be resolved before adoption, not after: exactly what permissions each agent identity holds, whether those permissions are audited the same way an employee's are, and what happens if a compromised agent uses its legitimate identity within the channel to escalate access or exfiltrate information. The productivity promise of integrating agents as "team members" is real, but it demands the same identity-management rigor already applied — or that should already be applied — to any account with access to internal systems.
Carlos Montiel is an enterprise AI solutions architect. He implements LLMs, Agents, RAG, and orchestrators for companies in Guatemala and Latin America. Contact him for a consultation.
Contact Carlos Montiel