Async Task Queue
for AI Agents
Reliable HTTP callback jobs with exponential backoff, priority queues, and dead-letter handling. Built for LangGraph, CrewAI, and any AI pipeline that needs guaranteed delivery.
# Enqueue a job in one API call
curl -X POST https://taskflowq.com/api/jobs \
-H "X-API-Key: tfq_your_key" \
-H "Content-Type: application/json" \
-d '{
"task_url": "https://yourapp.com/process",
"payload": {"prompt": "Summarize this doc"},
"max_tries": 5,
"priority": 80
}'
# → {"job_id": "01JOB...", "status": "pending"}
Everything your AI pipeline needs
Drop-in reliability layer between your AI agents and the tasks they trigger.
HTTP Callbacks
Your endpoint receives a POST when the job completes. No polling required.
Exponential Backoff
Failed jobs are retried automatically with configurable backoff and max attempts.
Priority Queues
Set job priority 0–100 to ensure critical AI workloads are processed first.
Dead-Letter Queue
Exhausted jobs land in a dead-letter queue for inspection and manual retry.
HMAC Security
Every callback is signed with a per-job HMAC secret. Verify before processing.
Tier-Based Rate Limits
1K/day on Free, 50K on Pro, 500K on Team. Upgrade without changing your code.
How It Works
Three steps from enqueue to completion.
Enqueue a Job
POST to /api/jobs with your task_url, payload, and optional priority or delay.
Worker Processes
Our worker picks up the job and POSTs to your task_url with the payload and HMAC signature.
Callback or Retry
Your service returns 200 to acknowledge. On failure, the job retries with exponential backoff.
Built for AI Agent Developers
From solo LLM experiments to production-grade AI pipelines.
Start building in 30 seconds
Register for free, get your API key, and enqueue your first job.
curl -X POST https://taskflowq.com/api/auth/register \
-H "Content-Type: application/json" \
-d '{"email": "[email protected]"}'
# → {"api_key": "tfq_...", "tier": "free"}