Generative AI & RAG
349 builders
Retrieval-augmented systems and LLM applications grounded in proprietary data.
The taxonomy
Filter the index on what actually matters.
Every AI builder in the Trustgent index is described against four axes: what they build, who they build it for, how they engage, and which regulatory regimes their delivery has been verified against. Each tile is a search facet, not a category wall.
Most buyers come to Trustgent with a known capability and a regulatory constraint, and need the intersection — “RAG builders verified for GDPR”, “document-understanding shops verified for HIPAA”. Click any tile to start the filter; combine more in the index header.
35 terms across four axes — growing as the index expands.
Capabilities
What providers build. Each capability is a class of system, not a tool. Click to see verified builders.
AI agents & automation
236 builders
Tool-using agents, workflow automation, agentic pipelines.
MLOps & evaluation
240 builders
Production pipelines, model evaluation, observability, drift detection.
Computer vision
126 builders
Visual perception systems — detection, segmentation, classification, OCR.
Document understanding
67 builders
Structured extraction from contracts, forms, reports; classification + routing.
Voice & conversational AI
39 builders
Speech-to-text, text-to-speech, conversational interfaces, telephony agents.
AI strategy & advisory
207 builders
Evaluation, roadmap, and procurement support — verified through shipped work, not decks.
Data engineering for AI
273 builders
Pipelines, feature stores, vector stores, retrieval infrastructure.
Custom model fine-tuning
41 builders
Domain-specific fine-tunes, LoRA / QLoRA, distillation.
AI governance & compliance
32 builders
Policy, risk, audit, regulatory implementation for AI deployments.
Capabilities are claimable at L1 and cross-referenced at L2. An AI-analyzed (L4) project tagged to a capability is what separates “they say they build it” from “they have shipped it”.
Industries served
Sectors providers have shipped to. Four sectors have dedicated buyer landings; the rest filter as search facets.
- Financial services221 providers
- Healthcare176 providers
- Retail & e-commerce111 providers
- Manufacturing & industrial130 providers
- Government & public sector65 providers
- Legal15 providers
- Media & publishing32 providers
- Energy & utilities54 providers
- Logistics & supply chain72 providers
- SaaS & technology19 providers
- Education31 providers
Engagement model
How providers contract for AI builds. The model usually correlates with budget band and timeline.
Regulatory readiness (verified)
Regulatory regimes against which a provider's delivery has been verified. Each flag is a Trustgent-verified attribute, not a provider self-claim.
- EU AI Actgrowing
- GDPRgrowing
- HIPAAgrowing
- SOC 2growing
- ISO 27001growing
- FedRAMPgrowing
- PCI DSSgrowing
- DORAgrowing
Regulatory flags are verified provider attributes, not self-claims. A provider must reach L2 (cross-referenced) or higher for a flag to display on their profile.
Browse by location
Filter the index by country or region.
Every provider is indexed to their headquarters country. Browse all verified builders in a market, or explore regional rollups spanning multiple countries.
Key markets
- United States81 providers
- United Kingdomgrowing
- India28 providers
- Germany19 providers
- France26 providers
- Netherlands17 providers
- Canada22 providers
- Australia10 providers
- Singapore4 providers
- Brazil3 providers
- Spain8 providers
- Poland13 providers
- Italy5 providers
- Sweden7 providers
- Switzerland7 providers
- Belgium17 providers
How we use this
The taxonomy is the filter. Verification is the ranking.
Once you filter the index on a capability, industry, or regulatory regime, providers are ranked by their highest verification level (L5 outcome-verified → L1 claimed) and recency. A free-plan L5 provider always ranks above a paid-plan L1 provider. Always.
