What is Donovan?

Donovan deploys specialized AI agents for mission‑critical workflows in government and defense environments. Its no‑code Agent Factory lets users build, customize, and connect agents to existing systems and data sources—including local uploads, cloud storage, and external databases such as GDELT—without developer intervention.

The platform enables evaluation and comparison of model performance across leading models, supporting DoD AI readiness goals and ensuring reliable, secure operation. Donovan operates on classified, air‑gapped networks and is cloud‑agnostic, running in Kubernetes containers for flexibility and scalability.

Traceability features provide transparent reasoning and guardrails that align with defense ethics principles, giving users confidence in the AI’s decisions. Integrated with SGP and prebuilt data connectors, Donovan delivers rapid, mission‑tailored AI solutions that are secure, compliant, and adaptable to specific operational parameters.

Donovan user reviews

Based on 28 reviews, 75.0% of users recommend Donovan, rated highly for value for money.

recommend

don't

28 reviews

Liked for

Worth the price 19 of 21

Quality results 12 of 21

Good integrations 10 of 21

Easy to use 9 of 21

All key features 9 of 21

Disliked for

Inconsistent results 6 of 7

Missing features 5 of 7

Not worth the price 2 of 7

Lacks integrations 2 of 7

Hard to use 1 of 7

Would you recommend Donovan?

Recommend this tool?

Donovan's key features

Build custom domain-specific agents
Train agents with reinforcement learning
Translate data to agent-ready formats
Red Teaming for vulnerability testing
Deploy fail-safe asynchronous agents
Evaluate performance with observability
Continuous improvement via human feedback

Donovan use cases

Design a secure, end‑to‑end threat intelligence pipeline that automatically pulls classified feeds, normalizes data, and triggers containment actions—all configured with Donovan’s no‑code Agent Builder and deployed on an air‑gapped Kubernetes cluster.
Build a rapid, defensively‑aligned incident response workflow that evaluates multiple LLMs for evidence triage, logs every reasoning step in a tamper‑proof trace, and delivers actionable reports to SOC analysts without any coding.
Create a cloud‑agnostic supply‑chain risk assessment tool that connects to proprietary data connectors, runs AI agents on classified infrastructure, and enforces strict defense guardrails while providing audit‑ready traceability for compliance.