Developer resources.

Machine-readable resources for agents and developers, and the research notes behind how we build.

AnovaGrowth Developer Resources

API contract
OpenAPI Specification
The machine-readable contract for documented operations, typed parameters, response schemas, and explicit authentication declarations.
Open the spec
Agent guidance
Agent Instructions
When to use AnovaGrowth, where to start, how to navigate the public resources, and which boundaries agents must respect.
Open the instructions
JSON manifest
Agent Manifest
A compact inventory of AnovaGrowth capabilities, services, and machine-readable resources.
Open the manifest
Markdown context
Site Guides
Concise and full markdown guides for retrieval systems that need company, service, and research context.
Open llms.txt

Custom Model Training: Methods; Fine-tuning methods to evaluate when a domain task and enough approved examples justify a custom model.
Agent Architecture: Patterns; Multi-step agent patterns with visible tool use, source context, human review, and clear ownership boundaries.
RAG Systems: Methods; Retrieval patterns that ground AI in approved company data and make the supporting context inspectable.
Voice AI: Research; Voice-agent research for phone interactions, including latency, escalation rules, consent, and review requirements.
Self-Hosted Inference: Deployment option; Open-weight deployment options for teams that need more direct control over privacy, latency, and cost.
Evaluation & Benchmarking: Methods; Evaluation methods for response quality, latency, cost, escalation, and observable business outcomes.

Support workflow: Reference pattern; A reviewed support pattern: retrieve approved context, draft a response, cite the source, and escalate edge cases.
Lead intake: Reference pattern; An intake pattern that captures the right context, prepares a CRM handoff, and leaves final qualification to the operator.
Document extraction: Reference pattern; A document-processing pattern that maps a source, validates fields, and routes uncertain records for review.
Voice intake: Research note; A research area for phone workflows where latency, disclosure, escalation, and conversational quality must be assessed together.

Start with the problem, not the model
We check whether AI is the right tool before building anything. Sometimes a well-designed workflow beats a model.
Use the best model for the job
Hosted and open-weight options are weighed against the task, the source boundary, operating cost and review needs before a choice is made.
Own your infrastructure
A deployment can include self-hosted inference when privacy, hardware, cost and ownership make that the defensible option.
Measure everything
Each system defines its checks, latency, cost, escalation and business signals before it is recommended for release.