Zespan is an AI agent observability and engineering platform. It traces every agent decision, tool call, handoff, and delegation in production. It also provides prompt versioning, built-in LLM-as-judge evaluations, guardrails, cost optimization, and an AI ops assistant called ZespanPilot.

How do I instrument my AI agent with Zespan?

Zespan requires 2 lines of code. Import zespan and call zespan.init({ apiKey: process.env.LT_KEY }). This auto-patches OpenAI, Anthropic, Gemini, Bedrock, and Mistral. For framework-level tracing, add one handler: ZespanCallbackHandler for LangChain, ZespanCrewAIListener for CrewAI, or ZespanADKHandler for Google ADK.

Does Zespan support prompt versioning?

Yes. Zespan includes prompt management with versioning, a playground for iteration, and A/B testing to compare prompt versions against each other in production.

What evaluations does Zespan support?

Zespan ships 12 built-in LLM-as-judge evaluation templates including faithfulness, relevance, toxicity, groundedness, and more. Evaluations run automatically on every trace with no custom scoring functions required.

How does Zespan compare to Langfuse?

Zespan is agent-native: every span carries agent identity, delegations are first-class trace events, and an agent map is built automatically. Langfuse was built for LLM pipelines and extended to agents later. Zespan also ships 12 built-in eval templates (Langfuse has none), includes an AI cost optimizer, and ZespanPilot for AI ops. Langfuse has open-source self-hosting; Zespan does not.

What is the free tier for Zespan?

The free tier includes 10,000 traces per month, 14-day retention, 2 projects, and 1 seat. No credit card required.

Haystack Integration

Native ZespanHaystackTracer for Haystack pipelines. Pass the tracer to pipeline.run() and every component execution, LLM call, and pipeline latency is captured — Python only.

Haystack (by deepset) is an open-source Python framework for building LLM-powered search and RAG pipelines with composable, production-ready components.

zespan.com — haystack trace

Zespan trace showing a Haystack pipeline run with component spans and LLM calls

Getting Started

Install the SDK

Install zespan alongside haystack-ai.

bash

pip install zespan haystack-ai

Add ZespanHaystackTracer to pipeline.run()

Initialize Zespan, create ZespanHaystackTracer, and pass it to your pipeline's run call. Your pipeline definition stays unchanged.

python

from zespan.sdk import Zespan
from zespan.integrations.haystack import ZespanHaystackTracer
from haystack import Pipeline
from haystack.components.builders import PromptBuilder
from haystack.components.generators import OpenAIGenerator

zespan = Zespan(api_key="your-api-key")
tracer = ZespanHaystackTracer(zespan)

pipeline = Pipeline()
pipeline.add_component("prompt_builder", PromptBuilder(template="Answer: {{question}}"))
pipeline.add_component("generator", OpenAIGenerator(model="gpt-4o"))
pipeline.connect("prompt_builder", "generator")

result = pipeline.run(
    {"prompt_builder": {"question": "What is Haystack?"}},
    tracer=tracer,
)

View traces in Zespan

Open Trace Explorer. Each pipeline run appears as a trace with spans for every component — prompt builders, generators, retrievers, and rankers.

What's captured automatically

Component execution spans: every pipeline component as a named span
LLM call tracing: token usage and cost attributed per generator component
Pipeline latency: total run time plus per-component breakdown
RAG pipeline support: retriever + generator spans linked in one trace
Python only: native Python tracer passed directly to pipeline.run()

FAQ

Do I need to modify my pipeline definition?

No. Add ZespanHaystackTracer only to the pipeline.run() call. Your Pipeline, component definitions, and connections stay unchanged.

Does this work with Haystack v2?

Yes. ZespanHaystackTracer targets Haystack v2's pipeline tracing interface.

Start for free — 10K traces/month, no card needed

Haystack integration works on all plans including the free tier.

Get started free →Get a demo

← All integrations