DEV Community: MW

The AI arms-race - a fallacy of cognition

MW — Tue, 02 Jun 2026 05:45:46 +0000

The arms-race for AI for the average person is pure fallacy. It's the cognitive load-creating friction in attempting to understand an extremely difficult subject completely deflating the subjcect-matter, despite never needing to understand its deepest tranches in the first place.

Example: you should always extract deterministic logic from a non-deterministic subject. I.e. use code (libraries and standard functions) for calculations, but use LLMs and their counterparts (HRMs, SLM, VLLMS, etc.) for their NLU abilities. Small models, whenorchestrated properly, can often outperform in terms of efficiency against architects who brute-force frontier-level intelligence.

Take our calculations, ephemeral subjects, and hard-maths from the model to be calculated by stable code written decades ago. Let the model itself determine what code to actually invoke.

Salesforce uses symbols like "->" for deterministic logic while using pipes like "|" for passing prompts to the LLM. This is basic orchestration framed in the Salesforce-method but shows up in different methods and manners from Flowise, Make.com, Zapier, AWS Step Functions, wtc. Yes, you need to understand some syntax but said syntax only appears like "magic" for a few weeks. Once you get your hands on it, the magic fades away and becomes just another tool in your belt.

Even the idea of embedded text is beyond the capabilities of most people, but this doesn't detract from delivering value. I don't get mad that a dog cannot use a calculator, nor that a squirrel doesn't understand biology. This is pure technological function applied to daily work tasks, and your ability to comprehend the underlying tech vs. your ability to explain it does not matter in day-to-day solving problems.

Are we really at an arms-race with the world? Or are we shadow-boxing foes that most likely we shall never meet? If you cannot communicate "temperature" to a product manager, how likely will it change a product when explaining cosine similarities in a geometric space to map semantic relationships gets lost in translation? While you, the architect and orchestrator must understand how these impact the product, this is not something that takes more than a few weeks of testing and QA to grasp its high-level affects. Furthermore, once you understand one set of models and weights, you are far more easily able to grasp other models and weights. Let Google, Anthropic, xAI, and the CCP fight out the "true" arms-race - the rest of us should be focused on building value without the fear of falling behind.

New Year, New You Portfolio Challenge - Evolving "Portfolios" into an Agent

MW — Fri, 30 Jan 2026 03:08:44 +0000

This is a submission for the New Year, New You Portfolio Challenge Presented by Google AI

About Me

I'm Michael Weed, a Solution Architect and Agentic AI Engineer who's spent the last decade bridging enterprise systems (Salesforce, AWS, Google Cloud) in high-compliance environments (HIPAA, PCI-DSS, SOC2). I've grown tired of static portfolios that say "I know AI" without proving it.

So I built Atlas-G Protocol, a portfolio that isn't just about AI - it is AI. Give it a try below.

Portfolio

How I Built It

The Core Idea is Agent-to-Agent Communication
Traditional portfolios are read by humans. Mine is designed to be queried by humans and agents. Built with the power of Google Antigravity to use Cloud Run functions.

The Atlas-G Protocol exposes an MCP (Model Context Protocol) server that allows any MCP-compatible environment - Claude Desktop, Gemini CLI, Cursor, or Antigravity - to connect and interview me programmatically. Your agent can ask my agent about my experience, verify my claims, and audit my project architectures, all through standardized tool calls.

This is the future of recruiting: Agents screening agents. Instead of a recruiter scanning a PDF, their AI assistant calls
mcp_query_resume("Experience with HIPAA-compliant systems")
and gets a verified, structured response.

Tech Stack

https://atlas-g-protocol-651633010401.us-central1.run.app/
The function is also deployed to full domain michaelweed.xyz

The Governance Layer

Every response passes through a validation layer that:

Cross-references claims against my resume knowledge graph
Blocks hallucinations before they reach the user
Logs compliance checks in real-time to a visible audit stream
If the agent tries to claim I worked somewhere I didn't, the Governance Layer catches it.

This is the kind of production-grade guardrail I architect for my clients, now demonstrated live.

What I'm Most Proud Of

The "Interview Me" MCP Server Other developers can point their AI tools at my portfolio and have a conversation with it. Here's how:

For Cursor / Claude Desktop / Antigravity users, the easiest way is to add it as a Native SSE server:

Type: SSE
URL: https://atlas-g-protocol-651633010401.us-central1.run.app/mcp/ Alternatively, use the mcp-remote bridge:

{
  "mcpServers": {
    "atlas-g-protocol": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://atlas-g-protocol-651633010401.a.run.app/mcp"]
    }
  }
}

Then your agent can call tools like:

mcp_query_resume("Salesforce experience")
→ Semantic search over my career
mcp_verify_employment("Salesforce", "Solution Architect")
→ Verify my claims
mcp_audit_project("Atlas Engine")
→ Deep-dive architecture with GitHub links
mcp_check_availability()
→ See if I'm available for contract work

The Live Audit Log
Watch the agent "think" in real-time. Every compliance check, every validation step, every governance decision is streamed to the UI. It's not a gimmick—it's proof of work.
Privacy by Design
My resume data never touches the public repo. The
resume.txt file is gitignored, and the template is provided for anyone who wants to fork and personalize. The agent reads from local data, so my PII stays private while my capabilities stay public.

Try it now: Connect your agent to my portfolio and see what agentic architecture actually looks like in production.

I Built a $150k AI Sales Acceleration Platform with AWS + Salesforce and Open-Sourced the Entire Thing

MW — Thu, 20 Nov 2025 11:42:55 +0000

Two weeks ago I finished a client-level project that cost $150k+ in consulting fees.

Then, I open-sourced every line of it.

Meet Atlas Engine → https://github.com/MichaelWeed/atlas-engine
Live demo → https://atlasengine.net (yes, it will actually call your phone)
In under 60 seconds it:

Captures a web lead
Creates a Salesforce Lead
Generates a personalized AI script with Claude 3.5
Calls their phone with Amazon Connect
Lets them talk naturally to the AI
Summarizes the call and updates Salesforce

100% free. 30-minute deploy. Go build something valuable