Loading
Tuhin Paul's contributions
Hybrid RAG with LangGraph: Vector, Keyword & Metadata Retrieval in Action
Master hybrid RAG with LangGraph: Combine vector, keyword, and metadata retrieval for production-ready, auditable AI answers.
Selecting the Right Embedding Model for a New Domain
Empirically select the best embedding model for specialized domains using LangGraph. Optimize RAG performance beyond general knowledge.
Measuring Retrieval vs. Generation Quality Independently using LangGraph
Master RAG evaluation by decoupling retrieval & generation with LangGraph. Pinpoint bottlenecks for production-grade AI.
Chunking Strategies for Enterprise RAG
Master enterprise RAG with advanced chunking strategies. Learn structure-aware and parent-child chunking for enhanced accuracy and context retention.
Building a Production-Grade RAG Pipeline: From Ingestion to Answer Generation with LangGraph
Build production-grade RAG with LangGraph. Go beyond basic retrieval with self-correcting loops for ingestion, query rewriting, and hallucination checks.
Redesigning Agentic RAG in LangGraph to Slash Latency and Cost
LangGraph is not just a tool for building loops; it is a state-machine orchestrator. By leveraging Model Cascading, Parallel Execution (asyncio/Send), and State Trimming, you transform a fragile, expensive demo into a robust, enterprise-grade production system.
Combining LangGraph and RAG for Enterprise Compliance
Building an AI agent for a regulated industry is an exercise in managing the tension between LLM flexibility and regulatory rigidity.
Planner-Executor vs. Graph-Based Orchestration (LangGraph) in Banking
By using LangGraph for the macro-orchestration, the bank satisfies regulators, ensures strict compliance, and enables seamless human-in-the-loop approvals. By nesting Planner-Executor agents inside specific nodes, the engineering team retains the flexibility needed to handle the messy, unpredictable nature of real-world customer documents.
Stop Using Mega-Prompts: Build a Multi-Agent Hospital for Your AI Using LangGraph
Build a multi-agent hospital for AI using LangGraph to process complex tasks efficiently, avoiding mega-prompt overload and hallucinations.
Building Robust E-commerce Agents with LangGraph and Pydantic
Build robust e-commerce agents with LangGraph & Pydantic. Prevent LLM hallucinations with tool grounding, structured outputs, and reflection.
Reducing Hallucinations in Agent-Based Systems Using LangGraph: An E-Commerce Order Booking
Combat LLM hallucinations in e-commerce with LangGraph. Learn techniques like grounded retrieval, schema enforcement, and self-correction for reliable order booking.
Measuring the Quality of Tool Selection by an AI Agent for a Credit Card System
Evaluate AI agent tool selection quality for credit card systems. Focus on process, not just outcome, for safety and efficiency.
Approval Layer for High-Risk LangGraph Actions in Enterprise Retail
Implement an AI Approval Layer in LangGraph for enterprise retail. Prevent catastrophic actions with human-in-the-loop for high-risk decisions.
Tracing Reasoning Failures Across Multiple Agent Steps in Retail
Discover how cascading reasoning failures in multi-agent AI systems can be traced and prevented in retail using shared state and reviewer agents.
Infinite Cycles in LangGraph Reasoning Chains
Master LangGraph's infinite cycles! Learn to prevent AI agent loops with state updates, attempt counters, and human escalation for robust reasoning.
Designing Guardrails for LangGraph Agents in Enterprise Retail
Secure LangGraph agents in retail with multi-layered guardrails: input validation, execution supervision, and output compliance for autonomous workflows.
Preventing Tool Over-Calling and Recursive Loops in Retail Agents
Prevent AI agent tool over-calling and recursive loops in retail with circuit breakers, state tracking, and strict tool definitions.
Designing Guardrails for Enterprise Agents with LangGraph in Retail
Secure enterprise AI agents with LangGraph. Implement robust guardrails for retail, preventing prompt injection, PII leakage, and unauthorized actions.
Designing Guardrails for AI Agents on Sensitive APIs and IoT Data
Secure AI agents on sensitive APIs & IoT data with multi-layered guardrails: IAM, DLP, action limits, and HITL for safe enterprise deployment.
Diagnosing a Critical AI Fraud Incident in UPI Payments
AI fraud detection failure in UPI payments during a flash sale, traced to a faulty data fallback mechanism. Learn how to prevent similar incidents.
Is Your LangGraph Agent Actually Moving the Needle, or Just Adding Complexity?
Evaluate whether LangGraph agents outperform simpler pipelines using business metrics, complexity costs, and a real-time news aggregator case study.
How to Prevent AI Agents from Over-Calling Tools
Prevent AI agents from over-calling tools with practical strategies like hard limits, state tracking, and optimized descriptions.
Multi-Agent Reasoning Loops: Failure Modes & Mitigations in One-tap UPI Payments
Explore failure modes in multi-agent AI systems, like deadlocks and hallucinations, and learn practical mitigation strategies for building robust AI applications.
Self-Correcting AI Agents Using Reflection Patterns in LangGraph
Build self-correcting AI agents with LangGraph's Reflection Pattern for robust, mission-critical applications like fintech.
Hybrid Retrieval (BM25 + Vector + Reranking) with LangGraph
Build a Hybrid RAG pipeline with LangGraph: BM25 + Vector Search + Reranking for efficient, parallel retrieval.
