Avesha
Also known as: Obliq, Avesha Systems
Autonomous AI SRE platform whose coordinated agent swarm monitors, diagnoses, and remediates issues across Kubernetes and cloud, with root cause analysis, auto remediation, and GPU orchestration.
Obliq is Avesha's autonomous AI SRE platform for production infrastructure across Kubernetes, virtual machines, serverless, bare metal, and AWS. A coordinated swarm of agents spanning SRE, QA, infrastructure, and data continuously collects metrics, logs, and events, isolates anomalies, and stitches telemetry into a causal narrative for root cause analysis in seconds. Reasoning agents weigh risk, cost, and policy, and an auto remediator executes or recommends validated rollbacks, replica bursts, and configuration hot patches, then documents the fix in Slack. The platform watches CI, pull requests, and generative AI commits to flag untested changes, and feedback loops retrain the system over time. Obliq installs into a customer's own cluster with Helm charts, integrates telemetry through MCP services for AWS, Prometheus, Loki, Neo4j, and CloudWatch, and adds an Elastic GPU Service for GPU orchestration, predictive scaling, and right sizing across multi cloud and on premises environments. It targets SRE and DevOps teams that want autonomous operations rather than reactive dashboards.
Vendor details
Canonical URL
https://avesha.io
Category
SRE / DevOps agent
Funding status
Venture backed
Company status
private
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Integrates telemetry through MCP services for AWS, Prometheus, Loki, Neo4j, and CloudWatch, plus Slack, DataDog, Kubernetes events, and OpenTelemetry; installs via Helm charts into the customer's cluster.
Sources & related URLs
Related / legacy domains
Capability coverage
10.5 / 14 capabilities · 75%
| Integrations & Tool Callingintegrates telemetry through MCP services for AWS, Prometheus, Loki, Neo4j, and CloudWatch, plus Slack, DataDog, and OpenTelemetry (docs.avesha.io) | Full |
|---|---|
| Workflow Orchestrationa coordinated swarm of SRE, QA, infrastructure, and data agents runs multi step root cause analysis and remediation with declarative playbooks (avesha.io) | Full |
| Knowledge Grounding & RAGgrounds analysis in telemetry, historical incidents, runbooks, and knowledge bases, stitching logs, traces, and metrics into a causal narrative (avesha.io) | Full |
| Human Oversight & Guardrailsthe auto remediator executes or recommends validated rollbacks and hot patches, with policy weighted decisions and declarative playbooks (avesha.io) | Partial |
| Security, Identity & Governanceinstalls self hosted in the customer's cluster with AWS IAM roles and slice aware isolation; specific certifications are not detailed (docs.avesha.io) | Partial |
| Observability & Auditabilitycontinuously collects metrics, logs, and events with OpenTelemetry and produces causal narratives, documenting fixes in Slack (avesha.io, docs.avesha.io) | Full |
| Memory & State Persistencetelemetry is stitched with memory across prompt, tool, and outcome, and feedback loops retrain the system over time (avesha.io) | Full |
| Deployment & Data Residencyinstalls into the customer's own Kubernetes cluster via Helm and runs across AWS, VMs, serverless, bare metal, multi cloud, and on premises (docs.avesha.io, avesha.io) | Full |
| Prebuilt Agents / Templates / Packsprebuilt SRE, QA, infrastructure, and data agents, an auto remediator, and declarative playbooks (avesha.io) | Full |
| Triggers & Channel Coveragewatches CI, pull requests, and generative AI commits, plus Kubernetes events and alerts, and monitors continuously (avesha.io) | Full |
| Model Flexibility & Routingcore AI services require an OpenAI API key; multi model routing is not documented (docs.avesha.io) | Unable to verify |
| APIs / SDKs / MCP Extensibilityextends through MCP services and Helm configurable services for integrations (docs.avesha.io) | Full |
| Testing, Debugging & Optimizationgenerates tests, flags untested diffs before rollout, and validates rollbacks, with reinforcement learning loops for self improvement; a formal agent evaluation framework is not detailed (avesha.io) | Partial |
| Browser / Computer-useoperates infrastructure through Kubernetes and cloud APIs; browser or computer use is not documented (avesha.io) | Unable to verify |
Pricing
Enterprise pricing; not publicly listed
Enterprise engagements scoped to environment size and workloads; self hosted install with provisioned credentials.
Cost watchouts
The agent requires the customer's own OpenAI API key, so model usage is billed separately; cost scales with monitored infrastructure and GPU workloads.
Variable cost rationale
Cost scales with cluster and environment size, monitored workloads, and GPU usage; the platform also requires the customer's own model API keys, adding usage cost. No public rate is available.
Sales call required
Yes — required for paid access
Free / trial
Credentials are provisioned through Avesha; no public self serve tier.
Lowest paid plan
Not publicly listed.
Key ambiguities
No public pricing; the self hosted model means model API usage and infrastructure cost sit on top of any license.
Related vendors
- Alertd — Agentic AI SRE and DevOps platform for AWS that analyzes cost,…
- 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…