Writing
Notes on AI engineering, RAG, MVP/POC builds & Python
Practical guides drawn from shipping RAG pipelines, AI agents, LLM-powered features, MVPs, and Python backends to production.
Why Your Contractor Business Has Outgrown Housecall Pro (And What to Do About It)
Five signs your contracting business has outgrown Housecall Pro or Jobber, the three paths forward, and why a custom operations platform can cost less over time than you think.
2026-09-23
How a Contractor Went From AR Spreadsheets to a Real-Time Payment Dashboard
A case study on replacing three accounts receivable spreadsheets with a real-time dashboard for a multi-division contractor: the problem, the build, and what changed.
2026-09-23
Building Permit Data as a Lead Source: How Contractors Find Jobs Before the Competition
How contractors can use public building permit data to find new construction projects 6 to 8 months before competitors, and how a permit pipeline turns raw data into scored, routed leads.
2026-09-23
AI Voice Agent Development: How to Build a Production AI Voice Agent in 2026
A practical guide to building a production AI voice agent — the STT → LLM → TTS architecture, latency budget, telephony integration, guardrails, and what it costs to build one.
2026-07-14
How to Reduce LLM API Costs in Production: 10 Proven Strategies
Ten practical strategies to cut LLM API costs in production — model routing, prompt caching, retrieval hygiene, output limits, and how to find where the spend actually goes.
2026-07-11
Why Your RAG System Isn't Accurate — and How to Fix Retrieval Quality
A practical troubleshooting guide for inaccurate RAG systems — diagnosing whether the problem is retrieval or generation, fixing chunking, reranking, and how to measure improvement.
2026-07-08
RAG POC Engineer: How to Scope and Ship a Retrieval-Augmented Generation Proof of Concept
A practical guide to scoping, building, and evaluating a RAG proof of concept — what a RAG POC engineer does, common pitfalls, and a realistic timeline.
2026-06-12
Hiring a Senior Python Developer for a Project: A Practical Checklist
What to look for when hiring a senior Python developer for project-based or contract work — technical depth, scope-fit, evaluation, and how to structure the engagement.
2026-06-12
LLM Integration for Production Apps: A Practical Guide for Engineering Teams
How to integrate LLMs (GPT-4, Azure OpenAI, LLaMA) into production applications — architecture patterns, RAG vs. fine-tuning, cost control, and MLOps for AI features.
2026-06-12
Principal AI Engineer for Hire: What the Role Actually Covers
What a Principal AI Engineer does differently from a mid-level ML or backend hire, when to bring one in on a project basis, and how to evaluate candidates for the role.
2026-06-22
MVP and POC Development for AI Startups: Scoping, Building, and Validating Fast
How to scope an AI MVP or POC so it actually validates your idea — what to build first, what to skip, typical timelines, and how to avoid the most common reasons AI MVPs stall.
2026-06-22
Hiring an AI Agent Developer: What It Takes to Build Production-Ready AI Agents
What an AI agent developer actually builds — tool use, memory, multi-step planning — common failure modes, and how to evaluate whether an agent is ready for production.
2026-06-22
Fractional AI for Startups: Engineer vs. Consultant vs. Advisor, and Which You Actually Need
The difference between a fractional AI engineer, consultant, and advisor for a startup, when each one beats a full-time hire, and how to structure the engagement so it ships.
2026-09-23
AI Chatbot Development Cost: What to Budget in 2026
A realistic breakdown of what AI chatbot development costs — by complexity tier, the line items that actually drive the number, and where teams overspend.
2026-06-22
Vector Database Comparison: Pinecone vs. pgvector vs. Weaviate for RAG
How Pinecone, pgvector, and Weaviate actually differ for production RAG — managed vs. self-hosted, cost at scale, and which one fits which kind of project.
2026-06-22
How Much Does an AI MVP Cost? A 2026 Budget Breakdown
What an AI MVP actually costs by scope tier, the cost drivers that matter most, and how to avoid the two most common budgeting mistakes founders make.
2026-06-22
Django vs. FastAPI for AI and LLM Applications: Which to Choose
How to choose between Django and FastAPI for an AI/LLM-powered application — async performance, ecosystem trade-offs, and when each one is the better fit.
2026-06-22
Hiring a Freelance AI Engineer on Upwork: A Practical Vetting Guide
How to write the job post, read a profile, and vet proposals when hiring a freelance AI engineer on Upwork — what actually predicts a good hire versus what's just noise.
2026-06-26
AI Agent Frameworks Compared: LangGraph vs. CrewAI vs. OpenAI Agents SDK
How LangGraph, CrewAI, and the OpenAI Agents SDK actually differ for production agents: control model, multi-agent support, and which one fits which kind of project.
2026-09-15
How to Build an MCP Server: A Practical Guide to the Model Context Protocol
What the Model Context Protocol is, how to build an MCP server that exposes tools, resources, and prompts, the stdio vs. HTTP transport choice, and the security that matters.
2026-09-16
How to Evaluate LLM Applications: Building Evals That Catch Regressions
A practical guide to evaluating LLM applications: building an eval set, deterministic checks vs. LLM-as-judge, validating your judge against humans, and running evals in CI.
2026-09-17