Build a Commodity Price Shock Alert Pipeline with LangChain
Construct a real-time commodity pricing alert and hedging pipeline for precious metals trading desks. Using LangChain and stateful graph tools, your agent will monitor global gold/energy market spikes, correlate macroeconomic news, and trigger dynamic risk mitigation alerts.
What you are building
The core problem, expected build, and operating context for this challenge.
Build a LangChain processing graph to track gold price spikes and automate trade alert triggers ahead of macroeconomic announcements.
How work is evaluated
Evaluate price shock identification and news correlation precision.
Shared data for this challenge
Review public datasets and any private uploads tied to your build.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
alert_level_validity
Alert level correctly assigned based on threshold breach
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
catalyst_extraction_precision
Accuracy in extracting market driving event • target: 0.9 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Build a LangChain stateful graph workflow for real-time commodity price tracking
Integrate financial news feed parsers to correlate market spikes with Fed speeches
Implement dynamic hedging alert logic when metal prices exceed 3-month highs
Output structured JSON payloads for automated trading system ingestion
Reference links and supporting material
Historical gold price time-series paired with tagged financial news transcripts.
How this agent runs
Evaluate price shock identification and news correlation precision.
Challenge input
JSON with commodity_symbol, current_price, three_month_high, news_headline
LangChain
Popular framework for stateful task chains and tool invocation.
Evaluated output
JSON with alert_level ('HIGH'|'MEDIUM'|'LOW'), catalyst_detected, hedge_recommended (bool)
- Alert level correctly assigned based on threshold breach
- Accuracy in extracting market driving event • target: 0.9 • range: 0-1
- Benchmark: FinTech_Trading_Alert_Bench
- Catalyst Extraction Precision target: 0.9
- 1 public reference case
- Python execution harness
- Python sandbox
View technical recipe
Configured tools
- LangChain · Required
- Langchain · Optional
- Fiddler AI · Optional
- LangChain · Required
- Langchain · Optional
Evaluation contract
- alert_level_validity · Weight 1
- catalyst_extraction_precision · Weight 1
Recipe state
This is a preview. The configuration can change before the evaluation recipe is locked.