Production AI Engineering
Ship AI systems that survive real users, real costs, and real failure modes. Architecture, evals, observability, governance, and the production tradeoffs most teams skip until it hurts.
Read moreFounder, Praxis FWD
Most AI projects die between demo and production.
I help leaders and their teams turn AI from experiments into results: systems that survive real users, workflows that change how work gets done, and decisions grounded in evidence. Client work runs through Praxis FWD. What I learn along the way, I write and build in public.
Book an intro call- 01Over 15 years building production software
- 020 → 50+ engineers scaled as CTO at Buoy Software
- 03FDA-cleared medical device software shipped
- 041+ yr daily Claude Code user, documented publicly
Latest insights
Practical writing and video on AI agent architecture, development workflows, and what actually holds up in production.
Read the latest
VideoHow Jev AI Works (and How to Use It in Production)
How Jev AI works: decision models return a probability for every allowed answer instead of generating text. Here's how they work and how to run one in production.
Using the OpenAI Decisions API to Decide When an Agent Speaks
How I decide when an AI agent replies in Slack: an OpenAI Decisions API probability, a threshold kept in code, and a RubyLLM eval that picks it.
Read moreMy Agentic Coding Workflow: Claude Code, Codex, and Pi
A year on from my Claude Code setup: how I run Claude Code, Codex, and Pi with herdr, and the shape, build, review, ship loop behind it.
Read moreOne Skills Library for Every Coding Agent
I merged claude-code-workflows and my old agent-skills repo into one library at dgalarza/agent-skills and removed the skills I no longer use. Here is the structure, the install changes, and the marketplace rename that breaks existing installs.
Read moreWebMCP: Make Your Site Agent-Ready Without a Separate MCP Server
WebMCP lets your web app register typed tools for browser agents with document.modelContext.registerTool. No separate MCP server. Here's how it works, and when you still need one.
Read morePartners & Clients
Areas of focus
Areas of focus
What I work on
Most of my time goes to a few problems: getting AI systems into production, designing agents that hold up, leading engineering teams through the change, and cleaning up what AI tools built in a hurry. I work on them with clients and write about them in public.
Production AI Engineering
Ship AI systems that survive real users, real costs, and real failure modes. Architecture, evals, observability, governance, and the production tradeoffs most teams skip until it hurts.
Read more
AI Agents
Design agents with memory, tools, and boundaries that actually hold up. Memory systems, orchestration, tool use, governance, and the production patterns behind agentic systems.
Read more
Fractional CTO
CTO-level engineering leadership, augmented by agents that extend what one person can ship — at a fraction of the cost of a full-time hire. Architecture, strategy, and shipped code.
Read more
Vibe Code Audit
A professional security and production-readiness review for apps built with AI coding tools. Written report with prioritized fixes, delivered in days.
Read more
Mastra Consulting
Building with Mastra? I help teams design, evaluate, and ship reliable TypeScript agent systems with architecture, evals, observability, workflows, RAG, memory, and implementation support.
Read moreWork with Praxis FWD
Most engagements start with a paid discovery, so you see evidence before you commit to a build. Teams that have already built something start with a Vibe Code Audit.
Book an intro call01
AI transformation
Find where AI makes the biggest difference in how your company works, then ship it.
02
AI executive coaching
Coaching for executives, applied to the work already on your desk.
03
Custom AI engineering
Build AI into the systems you already run.
04
AI-native engineering
Ship faster with AI, and catch mistakes before customers do.
From the clients
“It's rare to find an engineer with deep technical knowledge, strong business acumen, and who also speaks “human”...
Eric GolmanWareSpace, CMO
He turned out to be a true strategic partner who not only brought proper DevOps to that app and every other project we are working on at WareSpace, but also helped define the product direction, scope, roadmap, etc. through consultative working sessions.”
01 / 03
Case studies
Free resources
A few tools you can run against your own codebase right now — free, no call required. Start with where things actually stand before deciding what's worth fixing.
Read the latestToolToolTool
Codebase Readiness Assessment
A free agent skill that scores your repo across 8 dimensions for AI-agent readiness and gives you a prioritized improvement roadmap.
Agent Eval Scorecard
Score yourself across the four layers of agent evaluation in two minutes and find out where your quality gaps are before your users do.
Agent Governance Scorecard
30 yes/no questions across four dimensions. Find out if what you shipped holds up under scrutiny.
Ready to look at where your AI adoption stands?
30 minutes. No pitch deck. We'll talk through what you're working on and tell you honestly whether Praxis FWD can help.