R42 Institute · Cohort 9 · 2026

Ventora.
An AI team for
the solo founder.

A founder covers every function. Ventora brings specialist AI agents into one workspace, with shared context and human approval before anything goes out.

Ventora’s light-theme launch screen with Dex, Finn, Ivy, Mark, and PrueVentora’s dark-theme launch screen with its five agent characters
Ventora’s launch screen, with five distinct agent identities.
My role
Product design & frontend lead
Context
R42 Agentic AI Fellowship
Deliverable
Working Demo Day product
The brief

One founder.
Every function.

Switching between finance, marketing, fundraising, and product means switching tools and losing context. The team’s response was a small crew of agents with defined roles, one shared memory, and decisions gated by the founder.

What I owned

I led product design and frontend: the founder-facing experience, agent screens, responsive layouts, and light/dark interface treatments.

What we built together

A five-person fellowship team connected the interface to an agent framework and real tools. My design work sat alongside research, agent personas, and system integration.

4specialist roles
1routing agent
5fellowship teammates
01 / The product

A role for each agent.
A clear place to start.

Start with Dex when you don’t know whose desk a task belongs on, or go directly to a specialist. Each role has its own persona and access scope.

How the crew is organizedSelect a role
FounderBusiness context & final approval
Dex / Routing

The starting point when a founder does not know which specialist to ask. Dex routes the request to the relevant role.

Shared context. Scoped roles. Human approval.
A desk with context

Make the next
request easier.

Mark’s marketing desk offers prompts for leads and outreach. The suggestions are specific to the business, giving the founder an immediate way into the conversation.

Mark’s marketing desk with business-specific prompts for leads, follow-ups, and outreach
Original product screen: Mark’s desk and suggested requests.

From a request to a draft.

The marketing flow in the final demo.

  1. 01 / Ask

    “I need to find new wholesale customers for my store. Draft an outreach email to local cafes.”

  2. 02 / Hand off

    The interface identifies the specialist: “Mark took this from Dex.”

  3. 03 / Review

    Mark explains the limits: nothing sends until the founder approves, and real contact details are needed.

Original demo showing the handoff from Dex to Mark and Mark explaining that outreach remains a draft until approved
The handoff, specialist identity, and approval boundary stay visible in the conversation.

Watch the working demo.

Original Demo Day recording · 8 min

Recorded fellowship demo, presented by Mounica Policharla.Open demo in a new tab
02 / Design decisions

Make the system
easy to read.

The interface needed to explain who was working, carry the business context, and keep the founder’s authority clear.

01

Named roles and visible handoffs.

Dex, Finn, Ivy, Mark, and Prue have distinct identities. The conversation identifies when a task moves from the router to a specialist.

02

Shared context, surfaced in the UI.

Business-specific suggestions and a “From your crew” feed bring the agent team’s work back into the founder’s workspace.

03

Agents draft. The founder approves.

Client-facing output and irreversible actions require human review. The approval boundary is part of the product model, not an assumption left to the user.

The five-role navigation and From your crew feed, showing a scheduled competitor-analysis update
Agent navigation and a scheduled update in the original interface.
03 / Bringing it together

A product interface.
A working agent stack.

The team built three connected layers: the frontend the founder uses, the agents that route and draft, and the infrastructure that connects them to tools.

  1. 01 / Frontend

    Vercel UI

    The founder’s request goes to /api/agent.

  2. 02 / Routing

    Traefik / HTTPS

    Authentication and routing to the backend path.

  3. 03 / Agents

    Hermes profiles

    The gateway loads the relevant agent profile.

  4. 04 / Model & tools

    Claude + tools

    Scoped connections to Gmail, Sheets, Drive, and GitHub.

Where the engineering mattered

Keeping context, reaching real tools, and making the Vercel interface agree with the Hermes gateway were documented team challenges. The backend ran on Hostinger with Docker; agent permissions were scoped by role.

The team

Vihaan Malik
Product design & frontend
Abhijeet Chaudhari
Scope & user research
Arjun Dandekar
Agent framework
Tanya George
Agent personas
Mounica Policharla
System integration

R42 Institute · Led by Dr. Ronjon Nag · Mentor: Zixuan An

04 / Validation & next steps

A founder conversation
reshaped the product.

We worked directly with Praveena, founder of Soopra AI, to understand her operating challenges. Her input reframed proactive prompting, extended the specifications, and helped settle the human-approval policy.

At Demo Day

  • Lengthy response times
  • Limited live integrations
  • Manual approval for actions
  • Early agent-to-agent coordination

The next work

  • Get feedback from real users
  • Connect real business data
  • Test coordination between agents
  • Expand the crew and harden access controls
“He has demonstrated great business and product sense to design and build the project in the team from scratch.”
Zixuan An Fund Manager · R42 fellowship mentor
Source: the team’s final R42 Demo Day presentation.View the complete deck
Next case studyWellio