Fix mkdocs URLs and update blog post to current API (#41)

* Restructure Quick Start: lead with code + output, then explain

Show the ModelSelector call and results table first, then the
conceptual loop, then break down the four elements (agent, dataset,
eval function, models) separately. Add LLM-as-judge note.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* Add benchmark results table to Use Cases section

Show 20-100x cost savings across BFCL, HotpotQA, and MathQA
benchmarks with link to blog post.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* Fix mkdocs URLs and update blog post to current API

- Point site_url/repo_url to AgentOptimizer org instead of TianyiPeng
- Update blog post code example: agent_fn → agent, use ModelSelector

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

Author: TianyiPeng

docs/blog/posts/technical-deep-dive.md

@@ -198,10 +198,10 @@ pip install agentopt
 ```
 
 ```python
-from agentopt import ArmEliminationModelSelector
+from agentopt import ModelSelector
 
-selector = ArmEliminationModelSelector(
-    agent_fn=your_agent_factory,
+selector = ModelSelector(
+    agent=YourAgent,
     models={"planner": ["gpt-4o", "claude-sonnet-4-6"], "solver": ["gpt-4o-mini", "claude-haiku-4-5"]},
     eval_fn=lambda expected, actual: 1.0 if expected in str(actual) else 0.0,
     dataset=your_eval_dataset,

mkdocs.yml

@@ -1,8 +1,8 @@
 site_name: AgentOpt
-site_url: https://tianyipeng.github.io/agentopt
+site_url: https://agentoptimizer.github.io/agentopt
 site_description: Find the right LLM models for your AI agents
-repo_url: https://github.com/TianyiPeng/agentopt
-repo_name: TianyiPeng/agentopt
+repo_url: https://github.com/AgentOptimizer/agentopt
+repo_name: AgentOptimizer/agentopt
 
 theme:
   name: material