Field Notes — 32 Articles

Notes from the foundry

Engineering essays on generative AI, retrieval systems, and what it takes to ship intelligent software to production.

Browse by topic

Build an Agent with Human Approval Using AI SDK 6

Tutorial // AgentsBuild an Agent with Human Approval Using AI SDK 6Varun Raj Manoharan · 12 min read2026-06-23→02

Build a RAG Chatbot Over Your Docs with Claude and pgvector

Tutorial // RAGBuild a RAG Chatbot Over Your Docs with Claude and pgvectorVarun Raj Manoharan · 14 min read2026-06-23→03

Connect Your App to MCP Servers with the AI SDK

Tutorial // MCPConnect Your App to MCP Servers with the AI SDKVarun Raj Manoharan · 12 min read2026-06-22→04

Build an AI Agent with Tool Use Using the Claude API

Tutorial // AgentsBuild an AI Agent with Tool Use Using the Claude APIVarun Raj Manoharan · 13 min read2026-06-22→05

Build a Coding Agent with the Claude Agent SDK

Tutorial // AgentsBuild a Coding Agent with the Claude Agent SDKVarun Raj Manoharan · 13 min read2026-06-21→06

Build a Streaming AI Chat UI with the Vercel AI SDK

Tutorial // Frontend AIBuild a Streaming AI Chat UI with the Vercel AI SDKVarun Raj Manoharan · 12 min read2026-06-21→07

Build a Multi-File Refactor Agent with GPT-5.5

Tutorial // AgentsBuild a Multi-File Refactor Agent with GPT-5.5Varun Raj Manoharan · 13 min read2026-06-20→08

Build a PDF Data Extraction Pipeline with an LLM

Tutorial // Document AIBuild a PDF Data Extraction Pipeline with an LLMVarun Raj Manoharan · 13 min read2026-06-20→09

Spin Up Sandboxed Subagents with the Antigravity CLI

Tutorial // DevToolsSpin Up Sandboxed Subagents with the Antigravity CLIVarun Raj Manoharan · 11 min read2026-06-19→10

Build an MCP Server to Connect Claude to Your Database

Tutorial // MCPBuild an MCP Server to Connect Claude to Your DatabaseVarun Raj Manoharan · 13 min read2026-06-19→11

Self-Host an Open-Weight Coding LLM with vLLM

Tutorial // InfrastructureSelf-Host an Open-Weight Coding LLM with vLLMVarun Raj Manoharan · 12 min read2026-06-18→12

Build a Semantic Search Engine with Embeddings and Qdrant

Tutorial // SearchBuild a Semantic Search Engine with Embeddings and QdrantVarun Raj Manoharan · 12 min read2026-06-18→13

Shipping Generative AI Features Without Breaking Production

Engineering // LLMsShipping Generative AI Features Without Breaking ProductionVarun Raj Manoharan · 8 min read2026-06-18→14

Build an AI Voice Agent with Twilio and a Realtime Model

Tutorial // Voice AIBuild an AI Voice Agent with Twilio and a Realtime ModelVarun Raj Manoharan · 14 min read2026-06-17→15

Build an OAuth-Secured MCP Server

Tutorial // MCPBuild an OAuth-Secured MCP ServerVarun Raj Manoharan · 13 min read2026-06-17→16

Build a Durable Multi-Step Workflow with Claude Opus 4.8

Tutorial // WorkflowsBuild a Durable Multi-Step Workflow with Claude Opus 4.8Varun Raj Manoharan · 13 min read2026-06-16→17

Add an Eval Harness to Your LLM App

Tutorial // EvalsAdd an Eval Harness to Your LLM AppVarun Raj Manoharan · 12 min read2026-06-16→18

Build a Code Completion Backend with Codestral

Tutorial // DevToolsBuild a Code Completion Backend with CodestralVarun Raj Manoharan · 11 min read2026-06-15→19

Build a Multi-Agent Workflow with LangGraph

Tutorial // AgentsBuild a Multi-Agent Workflow with LangGraphVarun Raj Manoharan · 14 min read2026-06-15→20

Build an AI Code Review Bot for Your Pull Requests

Tutorial // DevToolsBuild an AI Code Review Bot for Your Pull RequestsVarun Raj Manoharan · 13 min read2026-06-14→21

RAG Is Not a Silver Bullet — It's a Retrieval Problem

AI // RetrievalRAG Is Not a Silver Bullet — It's a Retrieval ProblemVarun Raj Manoharan · 6 min read2026-05-29→22

How to Evaluate an AI Development Partner

Insights // HiringHow to Evaluate an AI Development PartnerVarun Raj Manoharan · 9 min read2026-05-22→23

Vendor Lock-In with LLMs: How to Keep Providers Swappable

Insights // ArchitectureVendor Lock-In with LLMs: How to Keep Providers SwappableVarun Raj Manoharan · 9 min read2026-05-15→24

Cutting LLM Cost 50% Without Wrecking Quality

Insights // CostCutting LLM Cost 50% Without Wrecking QualityVarun Raj Manoharan · 9 min read2026-05-08→25

What It Really Costs to Run an LLM Feature in Production

Insights // CostWhat It Really Costs to Run an LLM Feature in ProductionVarun Raj Manoharan · 10 min read2026-05-01→26

Should We Self-Host an LLM? A Cost and Control Framework

Insights // InfrastructureShould We Self-Host an LLM? A Cost and Control FrameworkVarun Raj Manoharan · 9 min read2026-04-24→27

RAG vs Fine-Tuning vs Long Context: Which, and When

Insights // ArchitectureRAG vs Fine-Tuning vs Long Context: Which, and WhenVarun Raj Manoharan · 10 min read2026-04-17→28

When AI Is the Wrong Tool (and Cheaper Options Win)

Insights // StrategyWhen AI Is the Wrong Tool (and Cheaper Options Win)Varun Raj Manoharan · 8 min read2026-04-10→29

The AI Readiness Checklist Before You Ship to Customers

Insights // ProductionThe AI Readiness Checklist Before You Ship to CustomersVarun Raj Manoharan · 9 min read2026-04-03→30

How to Scope an AI MVP That Won't Get Thrown Away

Insights // ProductHow to Scope an AI MVP That Won't Get Thrown AwayVarun Raj Manoharan · 9 min read2026-03-27→31

Estimating Timeline and Budget for an AI Build

Insights // PlanningEstimating Timeline and Budget for an AI BuildVarun Raj Manoharan · 9 min read2026-03-20→32

Measuring AI Quality: Evals Your Board Will Trust

Insights // EvalsMeasuring AI Quality: Evals Your Board Will TrustVarun Raj Manoharan · 9 min read2026-03-13→