Signals

Architectural patterns shaping the next generation of AI infrastructure

Analysis of 189 companies reveals conviction levels across 12 distinct build patterns.

Vertical Data Moats

high

Domain-specific datasets become the primary differentiator as foundation models commoditize. First-mover advantage in data accumulation creates durable competitive positions.

What This Enables

Unlocks AI applications in regulated industries (healthcare, finance) where generic models fail. Creates acquisition targets for incumbents.

Time Horizon0-12 months
Primary RiskData licensing costs may erode margins. Privacy regulations could limit data accumulation strategies.
Companies171

Notable: xAI, Skild AI, Playlist, Domyn, Etched.ai

Agentic Architectures

high

Autonomous systems that can plan, execute, and iterate without human intervention represent the next paradigm shift in enterprise software.

What This Enables

Enables full workflow automation across legal, finance, and operations. Creates new category of "AI employees" that handle complex multi-step tasks.

Time Horizon12-24 months
Primary RiskReliability concerns in high-stakes environments may slow enterprise adoption. Regulatory uncertainty around autonomous decision-making.
Companies124

Notable: xAI, Skild AI, HUMAIN, Domyn, humans&

Continuous-learning Flywheels

high

Products that improve from usage create compounding advantages. User data continuously refines model performance, increasing switching costs.

What This Enables

Winner-take-most dynamics in categories where this pattern is well-executed. Defensibility against well-funded competitors.

Time Horizon24+ months
Primary RiskRequires critical mass of users to generate meaningful signal. Privacy concerns may limit data collection.
Companies87

Notable: xAI, Skild AI, HUMAIN, Playlist, humans&

Micro-model Meshes

high

Orchestrating multiple specialized models outperforms monolithic approaches for complex tasks while reducing cost and latency.

What This Enables

Enables cost-effective AI deployment for mid-market. Creates opportunity for specialized model providers.

Time Horizon12-24 months
Primary RiskOrchestration complexity may outweigh benefits. Larger models may absorb specialized capabilities.
Companies84

Notable: xAI, Skild AI, Domyn, Etched.ai, humans&

Guardrail-as-LLM

high

Secondary AI systems that validate and filter primary model outputs address enterprise trust requirements.

What This Enables

Accelerates AI deployment in compliance-heavy industries. Creates new category of AI safety tooling.

Time Horizon0-12 months
Primary RiskAdds latency and cost to inference. May become integrated into foundation model providers.
Companies57

Notable: xAI, Playlist, Domyn, Parloa, OpenEvidence

RAG (Retrieval-Augmented Generation)

high

Grounding LLM outputs in retrieved facts addresses hallucination concerns and enables enterprise-grade accuracy for knowledge work.

What This Enables

Accelerates enterprise AI adoption by providing audit trails and source attribution. Reduces inference costs by minimizing context windows.

Time Horizon0-12 months
Primary RiskPattern becoming table stakes. Differentiation shifting to retrieval quality and domain expertise.
Companies50

Notable: xAI, Domyn, ClickHouse, Parloa, Baseten

Natural-Language-to-Code

medium

Emerging pattern in AI infrastructure.

What This Enables

Potential to unlock new application categories.

Time Horizon12-24 months
Primary RiskLimited data on long-term viability.
Companies20

Notable: HUMAIN, Emergent, Ivo, XBuild, Keyi Technology

Knowledge Graphs

medium

Emerging pattern in AI infrastructure.

What This Enables

Potential to unlock new application categories.

Time Horizon12-24 months
Primary RiskLimited data on long-term viability.
Companies17

Notable: xAI, Domyn, Ivo, Interos, RISA Labs