Alertd
Agentic AI SRE and DevOps platform for AWS that analyzes cost, security, compliance, and performance in plain English and generates human approved fix runbooks.
Alertd is an agentic AI SRE and DevOps platform for AWS cloud operations that runs specialized agents across cost, security, compliance, and performance. Teams query their AWS infrastructure in plain English and receive AI generated dashboards, evidence backed recommendations, and review and apply runbooks, with the platform auto discovering the fleet and auto classifying resources even when tagging is inconsistent. Alertd deploys read only within the customer's own Amazon VPC, integrates with Amazon Bedrock and customer selected LLMs, and is SOC 2 Type II certified. It stays read only and never changes anything itself; a human reviews and applies every fix through their own pipeline as Terraform, CloudFormation, or console steps. It also detects infrastructure change and drift, rating each change and grouping what matters into themes. Alertd was founded in 2024 by Geoff Hendrey and Freddy Mangum, veterans of Splunk, AppDynamics, and Cisco, and is available on AWS Marketplace with a free 30 day trial.
Vendor details
Canonical URL
https://www.alertd.ai
Category
SRE / DevOps agent
Funding status
Pre seed. Raised 3 million dollars in a pre seed round led by True Ventures. Launched out of stealth in November 2025.
Company status
private
Use cases & customers
Primary use cases
Target customers
Integrations
Deployed read only in the customer's own Amazon VPC, auto discovers the AWS fleet, integrates with Amazon Bedrock and customer selected LLMs, and generates Terraform, CloudFormation, and CLI runbooks for the customer's pipeline.
Sources & related URLs
Related / legacy domains
Research sources
Capability coverage
8.5 / 14 capabilities · 61%
| Integrations & Tool CallingIntegrates deeply with AWS, auto discovers the fleet, connects to Amazon Bedrock and customer selected LLMs, and generates Terraform, CloudFormation, and CLI runbooks for the customer's pipeline (alertd.ai, aws.amazon.com). | Full |
|---|---|
| Workflow OrchestrationSpecialized agents run analysis across cost, security, compliance, and performance and generate review and apply runbooks, but the platform stays read only and a human applies changes (alertd.ai, prnewswire.com). | Partial |
| Knowledge Grounding & RAGGrounds every answer in the customer's live AWS data with evidence such as the actual change diff and auto classifies the fleet; grounding in infrastructure data rather than a document RAG (alertd.ai). | Partial |
| Human Oversight & GuardrailsHuman in the loop by design: the platform is read only, generates runbooks, and a human approves and applies every change through their own pipeline (alertd.ai, prnewswire.com). | Full |
| Security, Identity & GovernanceSOC 2 Type II certified and deployed read only within the customer's own Amazon VPC for data ownership and control (aws.amazon.com, alertd.ai). | Full |
| Observability & AuditabilityDelivers AI generated dashboards, change and drift detection, and full visibility into AI reasoning and data sources for verifiable results (alertd.ai, prnewswire.com). | Full |
| Memory & State PersistenceRecords every change over time and lets teams share and reuse validated queries and workflows; persisted artifacts rather than agent memory (aws.amazon.com). | Partial |
| Deployment & Data ResidencyDeploys within the customer's own Amazon VPC on ECS or EKS so data stays in the customer account (aws.amazon.com). | Full |
| Prebuilt Agents / Templates / PacksShips specialized prebuilt agents for cost, security, compliance, and performance, though not a broader template library (prnewswire.com, alertd.ai). | Partial |
| Triggers & Channel CoverageRuns proactively and reactively, surfacing changes and drift as they happen, but is scoped to the AWS environment (prnewswire.com, alertd.ai). | Partial |
| Model Flexibility & RoutingLLM agnostic with a vendor neutral architecture supporting Amazon Bedrock plus customer selected models such as OpenAI, Anthropic, and Meta (aws.amazon.com, prnewswire.com). | Full |
| APIs / SDKs / MCP ExtensibilityNo public API, SDK, or MCP for building on the platform is documented; extensibility is described only as multi environment agnostic (prnewswire.com). | Unable to verify |
| Testing, Debugging & OptimizationToken economics and prompt inspection are internal cost controls; no user facing agent testing or evaluation tooling is documented (alertd.ai). | Unable to verify |
| Browser / Computer-useOperates through AWS APIs, not browser automation or computer use (alertd.ai). | Unable to verify |
Pricing
Free 30 day trial. Paid pricing is via AWS Marketplace and is not published; customers also bring their own LLM.
Platform subscription plus pass through model inference (bring your own LLM) and AWS infrastructure costs.
Included quota
Not public.
Cost watchouts
Bring your own LLM inference costs and AWS infrastructure costs in the customer's own VPC are on top of any platform fee.
Variable cost rationale
Customers bring their own LLM, so model inference is billed separately through Amazon Bedrock or another provider, and the platform runs in the customer's AWS account, adding AWS infrastructure cost; both scale with usage.
Overage / add-ons
Not public.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free 30 day trial.
Lowest paid plan
Not public.
Key ambiguities
Paid pricing is not published; only a free 30 day trial and AWS Marketplace availability are public.
Related vendors
- Avesha — Autonomous AI SRE platform whose coordinated agent swarm monitors,…
- Bluebricks — Context and control layer that lets AI agents operate cloud…
- Cased — AI infrastructure automation with pre configured agents, agent first…
- Causely — Causal reasoning engine that infers the single root cause of an…
- Chamber — AIOps agent for ML teams that puts GPU infrastructure on autopilot:…
- Cleric — Safety first AI SRE that investigates alerts read only and delivers…