The studio your own engineers build agents in.
Build RAG agents visually. Any model, any vector database, any knowledge graph. Governance built in.
Turning source documents into a governed knowledge base
Upload data. Upload documents, datasets, or structured data to be processed for information retrieval.
Everything needed to put an agent into production
Retrieval, redaction, evaluation, change control and cost governance — built once, shared by every agent.
Visual development studio
See the whole flow, stage by stage, instead of reading it in code.
API management
Every model connection, and every tool an agent may call, in one place.
Token and quota administration
Cost per user, team and agent — with quotas that throttle before they surprise you.
Model-agnostic by construction
Swap the model without rebuilding the agent.
Full software development lifecycle
Design, test, stage, ship — with change control and rollback.
The suite's compliance core
Every agent inherits the PHI boundary, access control and audit log.
Dev Studio is part of the InferAgents Suite, not a separate product, and shares its domain. It is the back end; Inferch is the front end your staff work in, and PayerAgents is the revenue-cycle product built on the same core.
What is Dev Studio?
Dev Studio is the back end of the InferAgents Suite — the environment where a health system's own engineering team builds AI agents. It provides API management, a visual development studio for retrieval-augmented generation agents, a full software development lifecycle from design through to production, and token and quota administration across every connected model.
Is Dev Studio sold separately from InferAgents?
No. Dev Studio is part of the InferAgents Suite and runs on the same HIPAA-compliant core as every other product in it. It is the suite's back end, not a standalone tool, and it shares the suite's domain at inferagents.ai.
Who uses Dev Studio?
Health-system IT, informatics and engineering teams. At University of Chicago Medicine the health system's own IT organisation builds agents in it directly, rather than filing requirements with a vendor and waiting.
Do you need to write code to build an agent?
Not for the common cases. Agents are composed visually from stages the platform already provides — ingestion, redaction, indexing, retrieval, checking — so a subject-matter expert can configure one. Connecting a new source system, scoping what a tool can reach, and approving a promotion to production remain engineering work.
How does Dev Studio handle tokens and cost?
Token and quota administration is built into the studio rather than reported after the fact. Usage and cost are visible per user, per team and per agent across every connected provider, with quotas that can throttle or route work when a threshold is reached.
What happens between building an agent and running it in production?
The full software development lifecycle: an agent is designed, tested against a repeatable evaluation set, staged, and promoted by an engineering owner. Every promotion is a change-controlled event with a version that can be rolled back.
Give your engineers the studio.
See how Dev Studio fits your model environment, your compliance requirements and the agents you already run.