One orchestrator.Every workflow.
Build, run, and govern across your entire organization.
Agentic AI under the same controls.
- Declarative
- Language-Agnostic
- 2,500+ Plugins
- Cloud
- Self-Hosted
- API-First
Kestra 2.0Available now!
Cross-Cloud ETL That Starts Itself
Every VM Configured, Every Run Traceable
Triage every issue from GitHub with AI, start each day with a plan
Self-service Business automation, fully recorded
Kestra scales because it is designed for cross-team orchestration from day one.
Adopt Once.
Standardize Everywhere.
Infrastructure Engineering
Automate Infrastructure With Governance, Not Scripts
Standardize Terraform, Ansible, CI/CD and operational workflows across hybrid and air-gapped environments.
6x
faster infrastructure delivery
90%
lower legacy tooling cost
trigger
Nightly AuditCheck Hosts System
Upload Output to S3
Send Slack Success
Data Orchestration
Ship Data Pipelines Faster and Keep them Reliable
Run ingestion, transformations, dbt, Airbyte, Spark and quality checks in one workflow engine.
10x
Faster Pipeline Delivery
90%
Fewer Manual backfills
dbt
dbt
Clone Repository
dbt Build
Python
Send Slack Notification
Send Email Notification
AI Automation
Operationalize AI Workflows Without Production Chaos
Agents, RAG pipelines, evaluation, publishing, retraining all orchestrated with governance and observability.
50x
less pipeline maintenance
3x
faster AI delivery cycles
trigger
Schedule monitorKubernetes deploy
AI inference
Python script
Slack notify
One Platform. All Your Workflows.
Declarative Orchestration
Write workflows in YAML. Version them. Review them. Ship them.
Connect Everything
Cloud, data, infra, CI/CD, ITSM, messaging. Plug into your entire stack.
2,500+ plugins
Batch and Event-Driven
Cron, events, webhooks, messages. One engine to run them all.
Governance and Reliability
Retries, timeouts, SLAs, RBAC, audit logs. SOC 2, Built to run in production.
Built for How Modern Teams Actually Work.
Any Language. No Lock-in
Run tasks in Python, Bash, Node.js, Go, or containers. No refactoring required.
Build Everything As Code & From the UI
Empower every team member to collaborate from a single tool—code for developers, UI for everyone else, all in sync.
Agentic Automation, Built In
Copilot for speed, agents for action. Create, iterate, and automate workflows with an AI native platform.
API First. CI/CD Native.
Trigger executions, manage workflows and resources through the API, and roll out updates safely with Git-driven CI/CD pipelines.
Work With Your Stack,
Not the Other Way Around.
Kestra integrates with your existing cloud, data, infrastructure, and SaaS tools.
With 2,500+ plugins, you can orchestrate everything without building custom glue code.
We processed billions of rows and thousands of weekly API pulls in under 3 months. For the first time, our analysts aren't waiting for engineering—they're building the workflows themselves.
- Billions
- of records processed securely
- < 3 Months
- to go from zero to full production
- 100+
- users empowered to build workflows
Deploy Anywhere. Keep Everything Governed.
Open Source
Self-host on Docker or Kubernetes. Free Forever
Enterprise Edition
Built for critical environments.
SSO, RBAC, audit logs, multi-tenancy. Isolated workers and dedicated task runners.
Hybrid environments: Cloud, On-Prem, Air-Gapped.
SLA-backed support and a dedicated customer success program.
Cloud
Managed platform, production-ready, scalable. Fastest time to value.
2230+ Blueprints, Ready to Build.
Jump-start your workflows with a growing library of workflow templates
ClickHouse Replica Lag and Readonly Health Gate
Classify a ReplicatedMergeTree table HEALTHY, AT_RISK, or UNREACHABLE from ClickHouse's own system.replicas, alert Slack, and optionally run SYSTEM RESTART REPLICA.
ChromaDB Vector Collection Snapshot and AWS S3 Disaster Recovery Pipeline
Automated ChromaDB collection preflight validation, paginated vector extraction to Parquet, and AWS S3 disaster recovery backup.
Chroma Vector Store Dangling Document and Quarantine Eviction Guard
Audit Chroma vector collections for tombstoned and dangling embeddings, archive quarantined documents to KV, safely evict stale vectors with human approval, and verify post-deletion state.
ClickHouse Detached Parts Cleanup Guard
Detect debris in ClickHouse system.detached_parts, drop the stale parts behind a dry-run and approval gate with a first-seen grace window, then re-read system.detached_parts and fail if any target remains.
ClickHouse Storage Disk Space and Capacity Headroom Sentinel
Audit ClickHouse system.disks storage utilization, detect volume exhaustion below 15% free, file GitHub issues, and alert Slack.
CKAN Open Data Incremental Sync to Parquet
Sync only the datasets that changed on a CKAN open-data portal since the last run, download their CSV files in parallel and convert each to Parquet with DuckDB, with a KV-stored cursor.
ClickHouse Mutation and Partition Parts Watchdog Sentinel
Monitor ClickHouse system.mutations and system.parts, detect stuck async mutations and partition parts explosion, and alert DBAs via Slack.
Monitor ClinVar genes for variant classification changes
Weekly NCBI E-utilities check of ClinVar for chosen genes. Keeps a snapshot in the KV store, compares recently modified records, runs a periodic full scan, and alerts Slack only on real changes.
CI/CD Failure Diagnosis for GitHub Actions and Kestra
Diagnose GitHub Actions and Kestra failures with log retrieval, deterministic error classification, sensitive-data redaction, and Markdown reports.
ClickHouse Cold Storage S3 Partition Tiering
Audit ClickHouse system.parts for partitions older than retention limits on local disk, recommending migration to S3 cold storage.
ClickHouse Parts Explosion Optimize Guard
Detect ClickHouse partitions whose active part count is climbing toward parts_to_throw_insert, merge them with OPTIMIZE TABLE PARTITION FINAL behind a dry-run and approval gate, then re-count active parts and fail if any target is still over the threshold.
ClickHouse Open Access Sentinel
Audit ClickHouse for accounts that are both passwordless and reachable from any host, alert Slack on breach, and optionally lock the host restriction down with ALTER USER.
Track ClinicalTrials.gov Study Status Changes, Results Postings and Completion Slips
Snapshot a ClinicalTrials.gov search and watched NCT ids every week, diff them in DuckDB and get a Slack digest of terminated or suspended trials, newly posted results, enrollment revisions and primary completion dates that slipped.
Citus Reference Table Replica Guard
Find Citus workers that are missing a reference table after citus_add_node, list the copies that would run, then call replicate_reference_tables behind an approval and fail the run if a worker is still short.
Cilium eBPF Map Pressure and Kernel Conntrack Saturation Sentinel
Monitors Cilium CNI eBPF map pressure to prevent kernel connection table overflow and pod packet dropouts.
ClickHouse Read-Only Replica Restore Guard
Find ReplicatedMergeTree tables stuck read-only because their ClickHouse Keeper metadata is gone, separate them from merely session-expired replicas, run SYSTEM RESTORE REPLICA behind a dry-run and approval gate, then re-read system.replicas and fail if any is still read-only.
ClickHouse Slow Query Threshold Gate
Classify ClickHouse query health HEALTHY, AT_RISK, or UNREACHABLE from system.query_log and system.processes, alert Slack, and optionally kill the longest-running active query.
Build a Chat-With-Your-Data System Using Elasticsearch and OpenAI (RAG)
Build a RAG chat-with-your-data flow that retrieves answers from Elasticsearch and generates grounded responses with OpenAI.
ClickHouse Stuck Mutation Killer Guard
Detect wedged ClickHouse mutations in system.mutations, kill the failing ones behind a dry-run and approval gate, then re-read system.mutations and fail if any targeted mutation is still pending.
CIS Control 1.6.4: Assess and Remediate /etc/motd Permissions over SSH
Assess CIS control 1.6.4 on /etc/motd over SSH against a KV-stored baseline and optionally auto-remediate ownership and permissions.
Kestra gives you fewer tools to maintain, fewer scripts to debug, fewer dependencies between teams, and one place to see what’s running
across your company.
One PlatformOne StandardOne Source of Truth