Vela – AI Scheduling Agents That Handle Multi-Party Chaos

TL;DR

TL;DR: Vela is an AI scheduling agent that reads context from emails, texts, and Slack, checks calendars, proposes times, follows up on ghosts, and rebooks when plans shift — handling multi-party, multi-channel scheduling so you never have to.

Source and Accuracy Notes

• Product: https://tryvela.ai
• HN Launch: Launch HN: Vela (YC W26)
• YC Profile: Y Combinator W26 batch

What Is Vela?

Scheduling is easy when it’s two people, one timezone, one channel. It becomes a constraint satisfaction problem when inputs are unstructured natural language across multiple communication channels, constraints change mid-solve, and the social dynamics don’t exist formally anywhere.

Vela is an AI agent that handles multi-party, multi-channel scheduling. You loop Vela into your emails, SMS, WhatsApp, Slack, phone or integrate into an ATS, and it takes over: reads context, checks calendars, proposes times, follows up when people ghost, and rebooks when things shift.

The core challenge Vela solves is the “one message, everything booked” scenario — where a recruiter sends one message and every interview across five candidates, three hiring managers, and two time zones gets booked, confirmed, and updated automatically. No links, no back-and-forth, no one spending hours with 20 emails.

Setup Workflow

Step 1: Connect Your Communication Channels

Vela integrates with multiple communication channels. The setup process involves connecting the channels your scheduling happens across:

# Vela supports:
# - Email (SMTP/IMAP integration)
# - SMS (via Twilio or similar)
# - WhatsApp
# - Slack
# - Phone (voice)
# - ATS integration (Greenhouse, Lever, Ashby)

Connect each channel through the Vela dashboard. Each channel requires standard OAuth or API credentials.

Step 2: Define Scheduling Constraints

Set up your availability windows, preferred meeting durations, buffer times between meetings, and timezone preferences:

{
  "availability": {
    "timezone": "America/New_York",
    "workHours": ["9:00", "18:00"],
    "bufferMinutes": 15,
    "maxMeetingsPerDay": 8
  },
  "meetingTypes": {
    "interview": { "duration": 45, "participants": ["candidate", "hiring_manager"] },
    "client_call": { "duration": 30, "participants": ["client", "account_manager"] }
  }
}

Step 3: Invite Participants

Forward or CC Vela on scheduling emails, or add participants through the Vela interface:

# Email Vela your scheduling request
vela schedule "Book interviews with candidates for Senior Engineer role"

Vela reads the context, extracts participant details, checks calendars, and proposes times.

Deeper Analysis

The Multi-Channel Identity Problem

When someone responds on SMS to a thread that started in email, Vela needs to unify identity, merge context, and continue without losing information. Phone numbers don’t map cleanly to emails, people use nicknames on text, and shared devices mean the responder might not be who you reached out to.

Vela solves this with probabilistic identity resolution across channels — matching names, numbers, and email addresses to build a unified participant profile.

Temporal NLU

“Next Friday” means different things on Monday versus Thursday. Vela extracts structured constraints from natural language and resolves them against calendar state. When ambiguity can’t be resolved automatically, Vela asks clarifying questions on the relevant channel.

Behavioral Datasets

The hardest part isn’t parsing — it’s applying the wrong interaction pattern for the wrong segment. C-suite folks respond to email within hours and expect formal3-option proposals. Truck drivers respond to SMS at odd hours from shared devices with “y tm wrks.” Vela has been building behavioral datasets from thousands of real interactions: response latency by role, channel preference by demographic, follow-up timing curves, and how many options to propose before hitting decision paralysis.

The Cascade Problem

When a client reschedules one interview, it cascades into four others. Vela tracks these dependencies and propagates changes across the entire schedule, notifying all relevant parties on their preferred channel.

Practical Evaluation Checklist

• [ ] Multi-channel integration works reliably (email, SMS, Slack)
• [ ] Identity resolution correctly merges participant profiles across channels
• [ ] Temporal NLU handles ambiguous date references
• [ ] Calendar conflict detection prevents double-booking
• [ ] Cascade updates propagate correctly when one meeting changes
• [ ] Ghost follow-up timing is appropriate for the use case
• [ ] ATS integration syncs candidate data correctly
• [ ] Privacy controls prevent Vela from accessing unintended calendar events

Security Notes

Vela processes sensitive scheduling data including participant contact information, meeting content, and calendar availability. Key security considerations:

• Vela should be granted read access only to relevant calendar events, not full calendar access
• Participant data should be retained only for the duration of the scheduling transaction
• Channel credentials (email, SMS) require OAuth with minimal scope
• Audit logs should track all scheduling decisions and channel accesses

FAQ

Q: What happens if Vela can’t find a suitable time slot?

A: Vela will present the closest available options with constraints highlighted — such as timezone conflicts or participant unavailability — and ask for guidance on which constraint to relax first.

Q: Can Vela handle timezone differences for global teams?

A: Yes. Vela resolves all times against participant-specific timezones and presents proposals in each participant’s local time. It accounts for business hours as defined per timezone.

Q: Does Vela work with existing calendar systems?

A: Vela integrates with Google Calendar, Outlook, and Apple Calendar via standard APIs. It also supports ATS systems like Greenhouse, Lever, and Ashby for recruiting workflows.

Q: How does Vela handle RSVPs on different channels?

A: Vela monitors all connected channels for participant responses, extracts the intent (accept/decline/reschedule), and updates the calendar accordingly. It reconciles conflicting responses across channels.

Q: What if a participant responds to the wrong thread?

A: Vela’s identity resolution can match responses to the correct participant even when they reply from a different email or phone number than the original invitation.

Conclusion

Vela tackles the unglamorous but universally painful problem of multi-party scheduling — the kind that generates hundreds of emails and zero productive work. By treating scheduling as a constraint satisfaction problem with real behavioral data, it removes the coordination overhead that eats into recruiters, executive assistants, and anyone who schedules across timezones and channels.

The real test is whether “one message, everything booked” actually works in production. Early customer results suggest it does — one staffing firm went from weeks of back-and-forth to fully booked in10 minutes of onboarding. If your workflow involves coordinating across multiple people, timezones, and communication channels, Vela is worth a look.

Try it at https://tryvela.ai.