Research
Deep research engine
Six providers, an 8-phase pipeline, automated prospect generation, and knowledge base enrichment.
Systems
Two research systems
User research agents
Configurable agents with custom schedules. Users define what to research, how often, and where to store results.
Internal KB enrichment
Automatic industry knowledge base building. SOP-driven queries aggregate and structure domain expertise.
Pipeline
8-phase research pipeline
1
Query formulation
SOP-driven, context-aware query generation.
2
Provider selection
Choose from 6 providers based on query type and quality tier.
3
Parallel execution
Fan-out to selected providers for concurrent research.
4
Result aggregation
Merge and deduplicate findings across providers.
5
Quality assessment
Confidence scoring and source verification.
6
Entity linking
Connect findings to existing CRM entities.
7
Knowledge storage
Vectorized storage with embeddings for semantic retrieval.
8
Synthesis
Actionable summary from all sources with recommendations.
Providers
6 research providers
| Provider | Specialty | Tier |
|---|---|---|
| Perplexity | General research, multi-turn queries | High |
| Firecrawl | Web scraping, structured extraction | High |
| Google Search | Broad web discovery | Medium |
| Company databases | Business intelligence, firmographics | Medium |
| Academic sources | Research papers, technical references | Specialized |
| News aggregators | Current events, industry trends | Medium |
Prospects
Prospect generation
14-step Inngest pipeline that researches a company, builds a knowledge base, and generates a server-rendered landing page.
POST /api/prospects/generate
|
1. Company research (Perplexity + web scraping)
2. LinkedIn scan (company + key people)
3. Knowledge base generation
4. Gap analysis (needs vs. capabilities)
5. Content generation (case studies + insights)
6. Page generation (SSR at /for/[slug])
|
Status: queued -> researching -> generating -> published
|
Output: Server-rendered prospect landing pageKnowledge
Knowledge base enrichment
SOP-driven templates
Structured query patterns for consistent, comprehensive knowledge gathering.
Vectorized storage
All research outputs embedded and stored for semantic retrieval.
Cross-entity sharing
Knowledge discovered for one entity enriches related entities and industries.
Automatic scheduling
KB enrichment runs on configurable schedules. Industry knowledge stays current.
0Providers
0Pipeline Phases
0Prospect Steps
Talk to Sentigen
Have questions about what you're reading? The intelligence behind this platform can walk you through it.