AI Coding
Analysis | Diagrams | Innovators
What is the best architecture for AI Coding and Agent Development?
Clarity of intents and purpose is now the most important capability for driving valuable outcomes. Code is only as good as it remains, understandable, maintainable and free of technical debt.
Build software to expand potential to deepen insights
Principles
Software development is now about guiding the intentions of intelligent systems through prompts. Those unwilling to adapt will fall far behind.
Flow of thought to meaningful action is the ultimate goal. Three key principles to mastering intent-driven software engineering.
- Precision Prompting
- Control Context
- Leverage Compute Power
Prompt Precision
Prompting is now programming, enabling multi-step workflows and dynamic problem-solving.
Balance prompt detail, with just enough context and the right LLM for the task at hand.
- Keep things simple, particularly when starting out
- Choose the right tools to constrain agent behavior
- Aim to only use prompts over writing code
- Plan out specs with prompts
- Factor out a library of reusable prompts
Optimize Context
Manage context to achieve results. Clarity of intentions is even more important with AI tools to communicate context, processes and desired outcomes to ensure solutions have a clean architecture and avoid wasting compute time and tokens.
- Product Requirements Document
- Diagraming Toolkit
- Clean Architecture
- Component Driven Design
- Software Algorithms
Needless context switching is the enemy of progress
Leverage Compute Power
Choose the right model for the job to be done.
Investing in advanced AI models yields significant productivity improvements compared to cost-saving approaches using lower-grade models.
Model Benchmarks
Use benchmarking tools to evaluate best bang for buck when choosing the best model for the job in hand.
Planning
The meta of the matter, matters most.
- Deep Wiki: Analyse github projects to understand how they work.
- Understand the Domain: Map the Data Footprint and Flow between entities that Transform Information and Potential into Meaningful Actions.
- Plan before you code: Identify the hardest problem to solve, clarify vision, architecture, and constraints in writing.
- Track tasks separately: Keep actionable work distinct from high-level planning.
- Document everything: Modular, persistent documentation helps maintain context and continuity.
- Enforce standards: Use configuration files to guide both human and automated contributors.
- Iterate and update: Continuously refine your plans and tasks as the project evolves.
Data Flow
At it's core software is all about the movement and transformation of data.
Data Flow: Understand how data flows through your system, how it created, stored, what impacts it's change of state, and who/what needs to know about that. Use flow diagrams to map the transformation of intent into valuable actions.
- Flow of Information: For information to be valuable it must be timely and actionable.
- Flow of Progress: The smooth, uninterrupted advancement of a project. Principles include clear process logic, synchronization, and minimizing waste. Practical steps to achieve this include defining clear steps and responsibilities and coordinating tasks and timelines.
- Flow of Value: The flow of value focuses on delivering maximum value to the customer with minimal waste. This involves value stream mapping, lean principles, and continuous improvement. Strategies include implementing lean methodologies and regularly assessing and improving processes.
What does the Optimum Toolkit for your Business Model look like?
Algorithms
What do you need your data for? What is the most complicated/valuable bit?
Make it work, make it right, make it fast. Use a spreadsheet to prove logic to get desired outcomes and document data flows.
Priorities
Functions and Features in order of delivery.
- Purpose: Tracks current tasks, backlog, and sub-tasks.
- Includes: Bullet list of active work, milestones, and anything discovered mid-process.
- Prompt to AI: “Update TASK.md to mark XYZ as done and add ABC as a new task.”
- Can prompt the LLM to automatically update and create tasks as well (through global rules).
AI Coding Tools
What are the best tools and practices for evolving solutions with a clean architecture?
| IDE | Notes |
|---|---|
| VS Code | Copilot |
| Aider | Most innovative? Python |
| Bolt New | Web Green Fields |
| Cline | VS Plugin |
| Convex Chef | Integrated Backend |
| Cursor | All Purpose |
| Google Firebase | Google Code Assist |
| Replit | Web Green Fields |
| v0 | Vercel, UI/UX Design |
| Windsurf | Open AI |
| coderabbit.ai | Dev Ops |
AI Agent Config
Prime your agents with purpose and rules.
| File/Directory | Purpose | Importance |
|---|---|---|
| .ide-project-rules | IDE and Project-specific standards & rules | Ensures consistency and compliance |
| Purpose | High-level vision & architecture | Guides all decisions, prevents drift |
| Priorities | Track tasks, backlog, and milestones | Maintains focus and progress |
| docs/plans/ | Modular planning documents | Supports detailed, persistent context |
| logs/tasks/ | Task history and rationale | Enables continuity and learning |
| README.md | Overview and instructions | Essential for onboarding and clarity |
MCP Config
What security checks need to be in place before using a MCP Service?
How do you find the perfect mix of MCP Servers for the task at hand?
See MCP Server Config Instructions.
Use Cases
Use JTBD Analysis to build software to expand potential to deepen insights to build closer connections.
- Save money on software with features you don't use or need
- Maximize Advantage of Trade Secrets and IP
- Take control of your data footprint
- Optimize integration and data flows
- Custom Deep Research
Prototyping Ideas
- Quickly building proof-of-concept applications
- Iterating on designs through conversation with AI
Automating Tasks
- Writing exploratory code to try out structures
- Simplifying and trimming down large codebases
- Automating monotonous tasks and one-off scripts
- Converting programs to more efficient languages for performance improvements
- Building entire web applications with unfamiliar technologies
Code Optimization and Refactoring
- Identifying opportunities to improve code efficiency
- Suggesting refactoring to simplify complex codebases
Documentation and Logic Explanations
- Generating code documentation
- Explaining complex code or algorithms
Context
Related principles and ideas.
- AI Prompts: Build a library of prompts tied to context for using them
- Model Providers: Base layer to innovate upon
- Augmented Workforce: The future of work
- AI Agent Frameworks: Build agents with deep domain knowledge
- Solana: Decentralized Compute at Lightspeed