Fix docs images not rendering

docs/active-learning-llm-oracle.md

@@ -5,7 +5,7 @@ description: Use everyrow's agent_map as an LLM oracle in an active learning loo
 
 # How to Replace Human Data Annotators with LLMs in Active Learning
 
-![Active Learning: Ground Truth vs LLM Oracle](images/learning_curve_accuracy.png)
+![Active Learning: Ground Truth vs LLM Oracle](/docs/images/learning_curve_accuracy.png)
 
 Human data labeling is slow and expensive. We replaced the human annotator with an LLM oracle in an active learning loop and achieved identical classifier performance — 200 labels in under 5 minutes for $0.26.
 

docs/scale-deduplication-20k-rows.md

@@ -7,7 +7,7 @@ description: Scale LLM-powered deduplication to 20,000 rows with linear cost, ac
 
 LLM-powered deduplication gives you semantic understanding that string matching can't, but naive pairwise comparison is quadratic. At 20,000 rows that's 200 million pairs. Everyrow's dedupe pipeline uses a funnel of embeddings, clustering, and targeted LLM calls to keep cost linear and accuracy high.
 
-![FDA Drug Products — Deduplication at Scale](images/fda_10pct_scaling.png)
+![FDA Drug Products — Deduplication at Scale](/docs/images/fda_10pct_scaling.png)
 
 Error rates stay near zero as scale increases. Cost and LLM calls scale linearly. Runtime is under 5 minutes up to 10,000 rows and 25 minutes at 20,000.