Series: Causal ML, LLM Fine-Tuning, and Agents

Ordered series for ML engineers: causal ML (SCM, IPS), LLM fine-tuning (SFT, LoRA, QLoRA, DPO), and production agentic systems (ColBERT, LangGraph, GraphRAG).

SERIES · 3 PARTS

Causal Machine Learning: Prediction to Intervention

SCM, backdoor adjustment, IPS, and MSM/IPTW for production decision systems.

  • P01Why Machine Learning Demands Causal Inference
  • P02Fixing Selection Bias in Enterprise Data with Inverse Propensity Scoring (IPS)
  • P03How to Solve Time-Varying Confounding & Mediator Bias in Longitudinal ML Pipelines
Open series
Causal Machine Learning: Prediction to Intervention

SERIES · 5 PARTS

LLM Fine-Tuning: SFT, LoRA, QLoRA & DPO

SFT, LoRA, QLoRA & DPO under VRAM and latency constraints.

  • P01The Definitive Guide to LLM Fine-Tuning: Objective, Mechanisms, and Hardware
  • P02Deconstructing LoRA: The Math and Mechanics of Low-Rank Adaptation
  • P03A Technical Guide to QLoRA and Memory-Efficient Fine-Tuning
  • P04Aligning LLMs with Direct Preference Optimization (DPO)
  • P05Building LoRA Multi-Adapter Inference on AWS SageMaker
Open series
LLM Fine-Tuning: SFT, LoRA, QLoRA & DPO

SERIES · 7 PARTS · 4 LABS

Production Agentic Systems

Shipping agentic systems under latency, cost, and governance constraints.

  • L01Autonomous Multi-Agent Network
  • L02Credit Underwriting Agent Systems
  • L03CodeContext: ColBERT Neural Code Search
  • L04Ad Agent on LangGraph CTR Loop
  • P01How to Design a Production-Ready RAG System (Architecture + Tradeoffs) (2026 Edition)
  • P02Understanding Vector Databases and Embedding Pipelines
  • P03How to Build Reliable RAG: A Deep Dive into 7 Failure Points and Evaluation Frameworks
Open series
Production Agentic Systems

Shipping AI Systems?

I help teams design and deploy scalable ML / RAG / LLM pipelines and MLOps infrastructure.



Or explore: