Global Financial Rewiring Trend Analysis with LangChain and Amazon Bedrock
Preparing strategy insights for the Singapore FinTech Festival 2026 requires synthesising macroeconomic trends across quantum security, AI, and cross-border digital finance. Implement a stateful LangChain LangGraph agent backed by Amazon Bedrock to aggregate financial intelligence, identify emerging macroeconomic risks, and compile structured executive briefings.
What you are building
The core problem, expected build, and operating context for this challenge.
Build a stateful multi-step LangChain agent using LangGraph and Amazon Bedrock (Claude models) to synthesize SFF 2026 global financial trend reports.
How work is evaluated
Evaluates state transition correctness, Bedrock model output coherence, and synthesis completeness of SFF 2026 strategic themes.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
graph_completion_check
Verifies that LangGraph executed all state transitions without termination errors
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
insight_coverage
Completeness of force analysis across target topics • target: 0.95 • 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
Construct stateful agent workflows using LangChain and LangGraph state machines
Integrate Amazon Bedrock model providers (Anthropic Claude 3.5 Sonnet / Bedrock Runtime)
Process five global forces data inputs: AI, Quantum, Web3, ESG, and Cross-Border Payments
Generate executive-ready briefing decks and risk assessment matrix reports
Reference links and supporting material
Curated collection of central bank announcements, SFF policy documents, and global fintech technology papers.
Challenge setup
Evaluates state transition correctness, Bedrock model output coherence, and synthesis completeness of SFF 2026 strategic themes.
Challenge input
JSON containing news feeds across 5 key financial forces
LangChain
Provides LangGraph state machine abstraction for multi-step intelligence workflows
Amazon Bedrock
Provides enterprise-grade, secure access to Claude 3.5 models on AWS
Evaluated output
JSON containing executive_summary, force_impact_matrix, and strategic_recommendations
- Verifies that LangGraph executed all state transitions without termination errors
- Completeness of force analysis across target topics • target: 0.95 • range: 0-1
- Benchmark: Macroeconomic Summarization Benchmark
- Insight Coverage target: 0.95
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
- LangChain · Required
- Langchain · Optional
- Amazon Bedrock · Optional
- LangChain · Required
- Langchain · Optional
Evaluation contract
- graph_completion_check · Weight 1
- insight_coverage · Weight 1
Metrics
- insight_coverage · Target 0.95
Benchmarks
- Macroeconomic Summarization Benchmark
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON containing news feeds across 5 key financial forces
Output: JSON containing executive_summary, force_impact_matrix, and strategic_recommendations
Execution harness: Python
Evaluation tools
- LangChain · Agent Framework
- Amazon Bedrock · Inference & Model Runtime
Recipe freeze marker
No recipe freeze time is recorded.