🤯
My AI Coding Workflow I spent the last week in a deep dive with Gemini 3 and Claude Opus. The goal wasn’t just to write code faster, but to find a workflow that handles complex architecture without falling apart.
If you’ve been following the rapid advancements in AI coding assistants, you know the landscape changes every week. This workflow leverages Gemini 3 and Claude Opus 4.5 primarily within Google Antigravity.
Why this stack? The Antigravity Pro plan offers generous limits that are hard to beat. To get comparable limits with Claude Opus directly, you’d likely be looking at a $100/month Claude code plan. But beyond the cost (though we all love saving money), it’s about using the right tool for the job.
I’ve developed a workflow that leverages the specific strengths of each model: Claude’s reasoning for architecture and Gemini’s massive context window for execution.
Orchestration
Most people just ask the AI to “write a snake game.” That works for demos. But for real-world, production-grade software, you need a pipeline.
My workflow looks like this:
graph TD
A[Start] --> B["Claude Opus 4.5: Research & Architecture"]
B --> C["Technical Document (~3800 lines)"]
C --> D[Split into Milestones]
D --> E["Claude Opus: Generate Prompts per Milestone"]
subgraph Execution Loop
E --> F["Gemini 3: First Pass (Code & Tests)"]
F --> G["Claude Opus: Review & Refine Code"]
G --> I["Claude Opus: Update Docs"]
I --> J["Gemini Flash: Commit Code"]
end
J --> K{"More Milestones?"}
K -- Yes --> L["Fresh Context Window"]
L --> E
K -- No --> M[Done]
Let’s break it down.
1. Architecture (Claude Opus)
I start with Claude Opus 4.5 to build a highly technical architecture document. This isn’t a brief summary; for example, in my recent project, this markdown file grew to around 3800 lines.
Pro Tip: Use “skills” for niche topics where general models might struggle or hallucinate.
For example, Swift concurrency is a fairly new topic, and very few people (or models) really know how to make it right. Injecting specific domain knowledge via skills ensures the foundation is solid before a single line of code is written.
2. Divide and conquer
You can’t just feed thousands of lines of specs into a coding task and expect perfection. I split this massive document into smaller milestone documents.
Crucially, each milestone doc contains all the necessary information for that step but zero noise related to future milestones. This keeps the model focused.
3. Prompt engineering
I don’t write the prompts for the coding agent. I ask Claude Opus to build the AI prompts for each milestone. I explicitly ask it to reference the correct detailed milestone doc we just created.
I feed it the prompt and the milestone doc and let it code. I also instruct it to write a lot of tests. The large context window means it can keep the entire current state of the project in “mind” without getting confused.
5. Review & Refine (Claude Opus)
This is the most critical step. On task completion, I don’t ask Gemini to fix its own bugs. I bring Claude Opus back in.
I ask Opus to review the code against the milestone and prompt. It finds the subtle bugs and architectural drifts. Crucially, I let Opus do the actual coding coverage to fix these issues and finalize the implementation.
6. Documentation & Commit
Once Claude finishes the coding, I have it update the documentation and any README files to reflect the new reality. Finally, I let Gemini Flash handle the git operations to commit the code.
7. Clean Slate
Then? I start a fresh conversation window for the next milestone. I give Gemini 3 the new prompt and the next milestone doc. This prevents context pollution from previous milestones and keeps the model focused on the task at hand.
Why this works
It turns coding into a management role. You are orchestrating expert agents:
- The Architect (Opus) ensures correctness and structure.
- The Engineer (Gemini) churns out volume and tests with deep context.
- The Reviewer (Opus) catches the mistakes.
If you are just using one model for everything, you are missing out on the synergy of their specialized capabilities. Give this workflow a shot on your next feature!
P.S. If you found this post a bit robotic, you’re not wrong. It was written by Claude. :D
I spent the last week in a deep dive with Gemini 3 and Claude Opus. The goal wasn’t just to write code faster, but to find a workflow that handles complex architecture without falling apart.
If you’ve been following the rapid advancements in AI coding assistants, you know the landscape changes every week. This workflow leverages Gemini 3 and Claude Opus 4.5 primarily within Google Antigravity.
Why this stack? The Antigravity Pro plan offers generous limits that are hard to beat. To get comparable limits with Claude Opus directly, you’d likely be looking at a $100/month Claude code plan. But beyond the cost (though we all love saving money), it’s about using the right tool for the job.
I’ve developed a workflow that leverages the specific strengths of each model: Claude’s reasoning for architecture and Gemini’s massive context window for execution.
Orchestration
Most people just ask the AI to “write a snake game.” That works for demos. But for real-world, production-grade software, you need a pipeline.
My workflow looks like this:
graph TD
A[Start] --> B["Claude Opus 4.5: Research & Architecture"]
B --> C["Technical Document (~3800 lines)"]
C --> D[Split into Milestones]
D --> E["Claude Opus: Generate Prompts per Milestone"]
subgraph Execution Loop
E --> F["Gemini 3: First Pass (Code & Tests)"]
F --> G["Claude Opus: Review & Refine Code"]
G --> I["Claude Opus: Update Docs"]
I --> J["Gemini Flash: Commit Code"]
end
J --> K{"More Milestones?"}
K -- Yes --> L["Fresh Context Window"]
L --> E
K -- No --> M[Done]
Let’s break it down.
1. Architecture (Claude Opus)
I start with Claude Opus 4.5 to build a highly technical architecture document. This isn’t a brief summary; for example, in my recent project, this markdown file grew to around 3800 lines.
Pro Tip: Use “skills” for niche topics where general models might struggle or hallucinate.
For example, Swift concurrency is a fairly new topic, and very few people (or models) really know how to make it right. Injecting specific domain knowledge via skills ensures the foundation is solid before a single line of code is written.
2. Divide and conquer
You can’t just feed thousands of lines of specs into a coding task and expect perfection. I split this massive document into smaller milestone documents.
Crucially, each milestone doc contains all the necessary information for that step but zero noise related to future milestones. This keeps the model focused.
3. Prompt engineering
I don’t write the prompts for the coding agent. I ask Claude Opus to build the AI prompts for each milestone. I explicitly ask it to reference the correct detailed milestone doc we just created.
I feed it the prompt and the milestone doc and let it code. I also instruct it to write a lot of tests. The large context window means it can keep the entire current state of the project in “mind” without getting confused.
5. Review & Refine (Claude Opus)
This is the most critical step. On task completion, I don’t ask Gemini to fix its own bugs. I bring Claude Opus back in.
I ask Opus to review the code against the milestone and prompt. It finds the subtle bugs and architectural drifts. Crucially, I let Opus do the actual coding coverage to fix these issues and finalize the implementation.
6. Documentation & Commit
Once Claude finishes the coding, I have it update the documentation and any README files to reflect the new reality. Finally, I let Gemini Flash handle the git operations to commit the code.
7. Clean Slate
Then? I start a fresh conversation window for the next milestone. I give Gemini 3 the new prompt and the next milestone doc. This prevents context pollution from previous milestones and keeps the model focused on the task at hand.
Why this works
It turns coding into a management role. You are orchestrating expert agents:
- The Architect (Opus) ensures correctness and structure.
- The Engineer (Gemini) churns out volume and tests with deep context.
- The Reviewer (Opus) catches the mistakes.
If you are just using one model for everything, you are missing out on the synergy of their specialized capabilities. Give this workflow a shot on your next feature!
P.S. If you found this post a bit robotic, you’re not wrong. It was written by Claude. :D