LangChain examples for Nebius Token Factory

This folder contains production-oriented LangChain examples powered by Nebius Token Factory. Each example is self-contained with its own requirements.txt, .env.example, source files, and sample data.

Examples

Example	Production use case	What it demonstrates
Voice Agent with Gradium	Conversational pitch and public-speaking coach	Browser voice turns, Gradium STT/TTS, LangChain prompt orchestration, Nebius reasoning
Incident Response Agent	SRE incident triage	Tool-driven log search, runbook lookup, deploy correlation, typed mitigation plan
Customer Support Resolution Agent	CX ticket resolution	KB search, order lookup, policy-grounded draft responses, approval-aware workflow recommendations
Vendor Risk Compliance Agent	Security/privacy vendor review	Policy control search, contract evidence review, data-residency checks, risk register output
Data Quality Ops Agent	Data operations investigation	Guarded read-only SQL, schema discovery, pipeline change correlation, reproducible data quality report

Setup pattern

Each project can be run independently:

cd langchain/<example-folder>
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
# add NEBIUS_API_KEY and any project-specific keys
python main.py

The voice agent also includes a FastAPI browser UI:

cd langchain/voice-agent-gradium-nebius-langchain
uvicorn voice_agent_server:app --host 127.0.0.1 --port 8501

Safety notes

All examples are designed to use local fixtures and read-only or draft-only tools.
No real infrastructure, refunds, shipments, or vendor approvals are executed.
.env.example files contain placeholders only; add real keys locally.