AI Developer Accelerator
Log In
Community
Classroom
Calendar
Members
Leaderboards
About
Log In
5
Patrick Chouinard
3d •
General discussion
AI Developer Accelerator — Coaching Call - July 7th
AI Developer Accelerator — Coaching Call - July 07
VIEW RECORDING - 86 mins (No highlights)
Meeting Purpose
A coaching call for AI developers to share project updates and strategies.
Key Takeaways
Fable's Strategic Value: The group is using Fable for high-level architecture and strategic analysis, reserving cheaper models like Opus for execution. This leverages Fable's superior "harness" (persistence and retry logic) for complex problem-solving.
System-Level Re-engineering: Patrick used Fable to analyze 160+ repositories and past chat logs, revealing hidden connections and generating a 3-month development pipeline. This included creating self-guided training modules and automated
install.md
files for internal services.
Predictive Maintenance: Ty developed "Q," a system that analyzes electrical signals to predict appliance failures before they occur. This enables proactive maintenance, reducing costs and downtime for service providers and their clients.
AI-Driven Demo Data: Paul used Fable to create virtual users on Android emulators, generating realistic demo data for his CRM. This solved a 3-year bottleneck, enabling immediate, high-quality product demonstrations.
Topics
Fable Usage & Strategy
High-Intensity Usage: Participants are maxing out Fable's weekly token limits, with the 50% promo extended to July 12.
Strategic Workflow: The group is adopting a multi-model pipeline:
GPT-5.5 Pro: Used for high-level problem framing and crafting detailed PRDs/prompts.
Fable: Used for complex architecture, strategic analysis, and PR reviews.
Opus/Sonnet: Used for code generation, executing Fable's plans.
Rationale: This workflow leverages Fable's superior "harness"—its ability to persist, retry, and iterate on complex problems—for high-value tasks, while using cheaper models for execution.
Project Updates
mdcatc:
Cemetery App: The app is ready, but the target county is delayed by a new mayor. mdcatc is now seeking a local sales rep to target other cemeteries.
Class to Curb: A logistics app for school carpool lines. A prior version reduced wait times from 2 hours to 15 minutes. The new version is on staging.
Ryan C:
Fable-Driven Development: Built multiple projects from scratch using Fable:
macOS finance app for automated accounting.
Estate agency website (launching next week).
Enhanced an existing estate agent CRM.
New Hardware: Purchased a Mac Mini to build a personal AI assistant for tasks like booking appointments.
Ty Wells:
Q (Predictive Maintenance): A system that analyzes appliance electrical signals to predict failures.
Mechanism: Establishes a 7-day baseline and monitors trends (e.g., wattage spikes during an ice harvest cycle) to detect deviations.
Goal: Enable service providers to identify failing parts, order them preemptively, and improve technician efficiency.
Market: Significant commercial potential (e.g., cold storage, vending machines, retirement communities).
Juan Torres:
AI Photo Booth: An app for real-time image transformation at events.
Progress: The multi-model AI pipeline is complete, with a system for adding new styles (e.g., "GTA 6").
Current Focus: Adding CloudWatch alerts for infrastructure monitoring (CPU, disk, memory) and documenting the physical setup/teardown process.
Paul Miller:
CRM Data Insights: Used Fable to analyze 10 years of CRM data, generating insights for management, sales teams, and customers.
Method: Built a deterministic data layer (SQL, LanceDB) for ground truth, using AI for interpretation and self-learning.
Automated Demo Data: Solved a 3-year bottleneck by using Fable to create virtual users on Android emulators.
Process: Fable controls multiple emulators, performs tasks, and generates realistic demo data, solving a key sales challenge.
Technical Deep Dive: Multi-Model Orchestration
Challenge: Efficiently manage token usage across multiple AI models.
Patrick's GitHub Workflow: Treats models as separate "people" using GitHub for handoffs.
Architect (Fable): Pushes specs to a repo.
Developer (Opus): Pulls specs, codes, and pushes.
QA (Hermes swarm): Tests the build and reports findings.
Ty's CMUX Automation Goal: Build a CMUX skill to automatically switch models based on token availability and task requirements.
Tool Mentioned:
omnigient.ai
was discussed as a potential model orchestration tool, though Ty noted it had memory issues in his brief test.
Next Steps
Patrick: Share the Fable prompt for system-level re-engineering on the forum.
Paul: Follow up with Ryan on the video for the NZ customer.
Juan: Research Prometheus and Alert Manager for centralized monitoring.
Ty: Investigate
omnigient.ai
and continue building the CMUX automation skill.
Action Items
Map VA cemeteries to private vs county-run; research county penalties -
WATCH (5 secs)
Build time-savings calculator for Class-to-Curb landing page -
WATCH (5 secs)
Deploy Class-to-Curb to production; notify Paul for school intros -
WATCH (5 secs)
Confirm Ryan's video received; if missing, request resend -
WATCH (5 secs)
Evaluate Prometheus + Alert Manager for centralized mini-PC monitoring -
WATCH (5 secs)
Post repo-analysis prompt on forum; DM Ty + Paul -
WATCH (5 secs)
Investigate
Omnigent.ai
for cross-model orchestration -
WATCH (5 secs)
Like
1
0 comments
5
AI Developer Accelerator — Coaching Call - July 7th
AI Developer Accelerator
skool.com/ai-developer-accelerator
Master AI & software development to build apps and unlock new income streams. Transform ideas into profits. 💡➕🤖➕👨💻🟰💰
11.3k
Members
67
Online
8
Admins
JOIN GROUP
Leaderboard (30-day)
1
Ty Wells
+9
2
Tom Welsh
+7
3
William Jin
+2
4
Elijah Stambaugh
+1
5
Keano Reurink
+1
See all leaderboards
Powered by