Loading
Tuhin Paul's contributions
- Article · Langchain
Transitioning from Stateless LangChain Prototypes to Secure, Stateful LangGraph Enterprise Solutions
Securely transition LangChain prototypes to stateful LangGraph enterprise solutions with custom tools for RBAC, API orchestration, and proprietary data.
- Article · AI Agents
The Challenge of Contradictory Evidence in Financial RAG
Navigating financial RAG contradictions: A hybrid framework for deterministic, LLM-driven, and human-in-the-loop resolution.
- Article · AI Agents
The Retrieval Strategy Verdict for Financial Anomaly Detection
Uncover the ultimate retrieval strategy for financial anomaly detection: Agentic Multi-Modal Retrieval, combining text, graph, and time-series data.
- Article · AI Agents
Metadata Architecture for Financial Vector Store Precision: A Six-Dimensional Framework for Enterprise RAG Systems
Unlock precision in financial RAG with a 6D metadata framework. Enhance retrieval, reduce hallucination, and ensure compliance.
- Article · Langchain
Automated Corporate Credit Covenant Monitoring
Automate corporate credit covenant monitoring with advanced RAG and multi-agent architecture, ensuring compliance and real-time risk assessment.
- Article · Langchain
Hybrid Chunking and Multi-Agent LangGraph for Financial Investigations
Hybrid chunking & multi-agent LangGraph for financial investigations, optimizing AML alerts with structured data and real-time context.
- Article · AI Agents
An Enterprise Regulatory Reporting System with Provenance-Chained Multi-Agent Architecture
Mitigating financial LLM hallucinations with a multi-agent architecture, provenance tracking, and layered defenses for regulatory reporting.
- Article · LLMs
An Enterprise AML Investigation System with Built-In Defenses
Discover an enterprise AML investigation system with built-in defenses against 14 LLM failure modes in financial reasoning.
- Article · Langchain
LLMs vs. Classical Models in Credit Scoring
LLMs enhance credit scoring by handling unstructured data, while classical models provide regulatory compliance and numerical accuracy. A hybrid approach ensures explainability and efficiency.
- Article · Langchain
LLMs vs Classical Models in a Deep Multi-Agent LangGraph Architecture with LLM Gateway
Hybrid LLM and classical ML architecture for enterprise AI, revolutionizing credit underwriting with RAG and multi-agent systems.
- Article · Langchain
RAG-Powered Credit Scoring vs. Traditional Feature-Based ML
RAG vs. traditional ML for credit scoring: explore LLM agents, heterogeneous data, and enhanced explainability for better underwriting.
- Video · Machine Learning
TensorFlow Tensors in Fintech
Explore how TensorFlow tensors are revolutionizing fintech, from fraud detection to algorithmic trading and risk management.
- Article · AI Agents
Building an Enterprise Multi-Agent ReAct Loop with Raw Ollama and LangGraph
Build enterprise multi-agent ReAct loops with raw Ollama & LangGraph for low latency, strict control, and persistent memory in regulated environments.
- Article · Langchain
Mastering Async Concurrency in Enterprise AI: gather vs as_completed vs TaskGroup in LangGraph
Master async Python for enterprise AI: gather, as_completed, TaskGroup in LangGraph for efficient ML inference and multi-agent concurrency.
- Article · Langchain
Custom Error Hierarchies in Enterprise LangGraph Multi-Agent Systems
Build resilient enterprise AI with custom error hierarchies in LangGraph. Achieve automated self-healing and enhanced observability.
- Article · Langchain
Solving the Imbalanced Data Problem in RAG
Solve RAG imbalanced data with pure Python weighted sampling. Ensure critical edge cases are represented, boosting LLM performance without external dependencies.
- Article · Langchain
Reproducible Deployments for Enterprise LangGraph Systems
Ensure enterprise LangGraph RAG systems are reproducible across environments with CI/CD, dependency locking, and dynamic state routing.
- Article · Langchain
Real-Time Fraud Detection Using Sliding Window Algorithms, LangGraph, RAG, State, and Memory
Build an enterprise fraud detection system using LangGraph, RAG, and sliding window anomaly detection for real-time financial transaction analysis.
- Article · Langchain
Automated Governance and Dynamic Routing in Enterprise LangGraph RAG
Automate RAG tool registration and enforce Pydantic schemas in enterprise LangGraph with Python metaclasses for scalable, secure multi-agent AI.
- Article · Langchain
Unit Testing Enterprise Multi-Agent LangGraph Systems
Master unit testing for enterprise LangGraph systems. Learn to mock async tools, LLMs, and test state persistence for robust AI deployments.
- Article · .NET Core
Dual-Layer Rate Limiting Middleware for Enterprise LangGraph RAG
Secure enterprise LangGraph RAG with dual-layer rate limiting (IP & API Key) in FastAPI, preventing abuse and controlling costs.
- Article · Langchain
Exponential Backoff, Jitter, and Multi-Agent LangGraph RAG
Implement robust retry logic with exponential backoff and jitter in Python for scalable, resilient multi-agent LangGraph RAG systems.
- Article · Python
Understanding __new__ vs __init__ in Python: Allocation, Initialization, and Object Pooling
Demystify Python's __new__ vs __init__: understand object creation, subclassing immutables, and implementing powerful patterns like Object Pooling.
- Article · Langchain
Mastering Messy Metadata: Flattening Nested Dicts in Enterprise LangGraph RAG
In enterprise AI, the quality of your RAG system is strictly bounded by the quality of your data pipeline. By implementing a collision-aware flattening utility and wrapping it in a strictly layered, multi-agent LangGraph architecture, you ensure that messy, real-world enterprise data is normalized without silent data loss.
- Article · Langchain
Architecting Enterprise Multi-Agent Systems: Solving Circular Imports in LangGraph RAG
Learn how to prevent circular imports in Python LangGraph RAG systems using dependency inversion, dependency injection, layered architecture, and decoupled tools.
