AI SalesConceptProspecta
An AI sales assistant that qualifies, briefs and drafts in seconds.
An AI assistant for inbound sales: scores every lead against your ideal customer profile, researches the account, briefs the rep and drafts a first reply for approval.
- Category
- AI · Sales Automation
- Designed for
- Web
Why Prospecta exists
The problem
Inbound leads wait hours for a reply, good ones go cold, and reps waste time researching people who were never a fit.
The product
Prospecta scores and researches every lead in seconds, briefs the rep and drafts a first reply that a person approves.
Built for: B2B sales teams receiving 100 to 10,000 inbound leads a month.
- Fit scoring with the reasons behind every score
- Account research from public sources, cited
- Rep briefs with suggested discovery questions
- Drafted first replies that a rep approves
- Follow-up steps that stop politely
- Two-way CRM sync and meeting booking
Walkthrough
Fast, Relevant, Still Human
Designed so speed never costs quality: the AI prepares, the rep decides and sends.
- 01
Inbound Leads
Every new lead with fit score, the reasons behind it, owner and status, plus daily totals.
- 02
Lead Brief
What they asked, what we know from public sources, likely priorities, fit breakdown and discovery questions.
- 03
Drafts To Review
A personalised first reply and follow-up plan for the rep to edit, approve and send.
- 04
Performance
Response time, meetings booked, show rate, pipeline created and how scores relate to outcomes.
Every screen
The full experience.
What makes it different
Explainable Qualification
Every score shows the criteria behind it and every research claim links to its public source.
Your Criteria, Written Down
Qualification rules and examples you control, applied consistently to every lead.
Cited Research
Company facts come from public sources that are linked in the brief.
Human Approval
Reps approve every first message; nothing is sent blind.
CRM-Native
Scores, briefs and activity sync to the CRM the team already uses.
How it’s built
The technology stack.
The architecture we would build this on today, and why each piece earns its place.
- Web app
Next.js, React, TypeScript, Tailwind CSS
Review queues and dashboards that stay fast with thousands of items
- AI services
Python (FastAPI) · Claude via the Anthropic API with tool use
Strong reasoning and structured tool calls, isolated from the core API
- Tools & integrations
Model Context Protocol (MCP) servers
One standard way for agents to reach business systems, with scoped permissions
- Durable workflows
Temporal
Long-running, multi-step AI work that survives restarts and waits for approvals
- Knowledge & memory
PostgreSQL + pgvector
Retrieval over approved sources without a separate vector database
- CRM
HubSpot and Salesforce APIs
Two-way sync so reps stay in their CRM
- Evaluation & tracing
Versioned eval suites, OpenTelemetry traces, cost tracking
Every prompt or model change is tested before release and every run is explainable
More inside
From Form Fill To First Meeting.
The inbound workflow, minutes instead of hours.
Meeting Booking
Proposed times from the rep's calendar with automatic holds.
Compliance
Consent and opt-out handling built into every sequence.
What comes next
On the roadmap.
What a second release would add once the first one is in real users’ hands.
- 01Website visitor identification
- 02Chat qualification on the website
- 03Territory and round-robin routing rules
- 04Call notes to CRM
- 05Win/loss feedback into scoring
Related
Services and industries behind this work.
Concept project
Need Something Like This?
This is concept work: the product design, screens and architecture we would start from. If your business needs AI-assisted sales workflows connected to your CRM, we can scope a first release with you, usually within a week.
Next project
Campusly
One platform for classes, coursework, attendance and fees.
