Log conversations

You can log chat conversations to the Opik platform and track the full conversations your users are having with your chatbot. Threads allow you to group related traces together, creating a conversational flow that makes it easy to review multi-turn interactions and track user sessions.

Understanding Threads

Threads in Opik are collections of traces that are grouped together using a unique thread_id. This is particularly useful for:

• Multi-turn conversations: Track complete chat sessions between users and AI assistants
• User sessions: Group all interactions from a single user session
• Conversational agents: Follow the flow of agent interactions and tool usage
• Workflow tracking: Monitor complex workflows that span multiple function calls

The thread_id is a user-defined identifier that must be unique per project. All traces with the same thread_id will be grouped together and displayed as a single conversation thread in the Opik UI.

Logging conversations

You can log chat conversations by specifying the thread_id parameter when using either the low level SDK, Python decorators, or integration libraries:

1
import { Opik } from "opik";
2
3
const client = new Opik({
4
apiUrl: "https://www.comet.com/opik/api", // Only required if you are using Opik Cloud
5
apiKey: "your-api-key",
6
projectName: "your-project-name",
7
workspaceName: "your-workspace-name", // Only required if you are using Opik Cloud
8
});
9
10
const threadId = "your-thread-id"; // any unique string per conversation
11
12
// Option A: set on trace creation
13
const trace = client.trace({
14
    name: "chat turn",
15
    input: { user: "Hi there" },
16
    output: { assistant: "Hello!" },
17
    threadId
18
});

Thread ID Best Practices

Generating Thread IDs

Choose a thread ID strategy that fits your application:

1
import uuid
2
import opik
3
4
# Generate unique thread ID per user session
5
user_id = "user_12345"
6
session_start_time = "2024-01-15T10:30:00Z"
7
thread_id = f"{user_id}-{session_start_time}"
8
9
@opik.track
10
def process_user_message(message, user_id):
11
    return "Response to: " + message
12
13
process_user_message("What is Opik ?", opik_args={"trace": {"thread_id": thread_id}})

Integration-Specific Threading

Different integrations handle thread IDs in various ways:

1
from opik.integrations.langchain import OpikTracer
2
3
# Set thread_id at tracer level - applies to all traces
4
opik_tracer = OpikTracer(
5
    project_name="my-chatbot",
6
    thread_id="conversation-123"
7
)
8
9
# Or pass dynamically via metadata
10
chain.invoke(
11
    {"input": "Hello"},
12
    config={
13
        "callbacks": [opik_tracer],
14
        "metadata": {"thread_id": "dynamic-conversation-456"}
15
    }
16
)

Reviewing conversations

Conversations can be viewed at a project level in the threads tab. All conversations are tracked and by clicking on the thread ID you will be able to view the full conversation.

The thread view supports markdown making it easier for you to review the content that was returned to the user. If you would like to dig in deeper, you can click on the View trace button to deepdive into how the AI assistant response was generated.

By clicking on the thumbs up or thumbs down icons, you can quickly rate the AI assistant response. This feedback score will be logged and associated to the relevant trace. By switching to the trace view, you can review the full trace as well as add additional feedback scores through the annotation functionality.

Scoring conversations

It is possible to assign conversation level feedback scores. For that, you need to understand how threads work in Opik. Threads are aggregated traces that are created when tracking agents or simply traces interconnected by a thread_id. In order to score a conversation, we need to ensure that the thread is inactive, meaning that no new traces are being created.

Threads are automatically marked as inactive after the timeout period and you can also manually mark a thread as inactive via UI using the Status button on top right corner of the thread view.

Once a thread is inactive, you can assign a feedback score to the thread. This score will be associated to the thread and will be displayed in the thread view.

And in the conversation list, you can see the feedback score associated to the thread.

You can also tag a thread and add comments to it. This is useful to add additional context during the review process or investigate a specific conversation.

Advanced Thread Features

Filtering and Searching Threads

You can filter threads using the thread_id field in various Opik features:

In Data Export

When exporting data, you can filter by thread_id using these operators:

• = (equals), != (not equals)
• contains, not_contains
• starts_with, ends_with
• >, < (lexicographic comparison)

In Thread Evaluation

You can evaluate entire conversation threads using the thread evaluation features. This is particularly useful for:

• Conversation quality assessment
• Multi-turn coherence evaluation
• User satisfaction scoring across complete interactions

Thread Lifecycle Management

Threads have a lifecycle that affects how you can interact with them:

1. Active: New traces can be added to the thread
2. Inactive: No new traces can be added, thread can be scored

Programmatic Thread Management

You can also manage threads programmatically using the Opik SDK:

1
import opik
2
3
# Initialize client
4
client = opik.Opik()
5
6
# Search for threads by various criteria
7
threads = client.search_traces(
8
    project_name="my-chatbot",
9
    filter_string='thread_id contains "user-session"'
10
)
11
12
# Get specific thread content
13
for trace in threads:
14
    if trace.thread_id:
15
        thread_content = client.get_trace_content(trace.id)
16
        print(f"Thread: {trace.thread_id}")
17
        print(f"Input: {thread_content.input}")
18
        print(f"Output: {thread_content.output}")
19
20
# Add feedback scores to thread traces
21
for trace in threads:
22
    trace.log_feedback_score(
23
        name="conversation_quality",
24
        value=0.8,
25
        reason="Good multi-turn conversation flow"
26
    )

Next steps

Once you have added observability to your multi-turn agent, why not:

1. Run offline multi-turn conversation evaluation
2. Create online evaluation rules to score your multi-turn conversations in production