Hi, I'm Rohit Raj
Senior AI Engineer crafting intelligent systems at the intersection of large language models, agentic workflows, and financial technology.
Focus Areas
Deep expertise across the AI/ML stack, from research to production systems.
LLM Systems
Production LLM architectures, inference optimization, and scalable deployment patterns.
Agentic Workflows
Autonomous AI agents, tool-use orchestration, and multi-agent collaboration for enterprise.
RAG & Retrieval
Advanced retrieval-augmented generation, vector databases, and semantic search pipelines.
ML Infrastructure
MLOps, model serving, feature stores, and end-to-end training pipelines at scale.
NLP & Language
Transformer architectures, fine-tuning strategies, and domain-specific language models.
Fintech AI
AI applications in financial services: risk modeling, compliance, fraud detection.
Latest Posts
Thoughts on AI, engineering, and building at scale.
Agentic Workflows in Fintech: Orchestrating LLM Agents for Autonomous Decision-Making
How autonomous AI agents are transforming financial services — from loan underwriting to fraud investigation — and the architectural patterns that make them production-ready.
Prompt Engineering for Production: Beyond Basic Prompts
Moving past toy prompts — a systematic guide to prompt design patterns, reliability techniques, and testing strategies for production LLM applications.
Vector Databases Demystified: Choosing the Right One for Your AI Stack
A practical comparison of Pinecone, Weaviate, Qdrant, pgvector, and Chroma — covering indexing algorithms, performance tradeoffs, and when to use each.
AI Research & LLM Insights
Deep dives into production AI systems, model architectures, and lessons from deploying LLMs in regulated industries.
Agentic Architectures
How autonomous AI agents are reshaping enterprise workflows in financial services.
Inference at Scale
Techniques for optimizing LLM inference: quantization, batching, speculative decoding.
Retrieval-Augmented Generation
Building RAG systems that actually work: chunking, reranking, and evaluation.
Evaluation & Safety
Frameworks for evaluating LLM outputs in high-stakes financial applications.