diff --git a/05_scheduling/hackernews_alerts.py b/05_scheduling/hackernews_alerts.py
index 357bc8c3d..f6e7bdee6 100644
--- a/05_scheduling/hackernews_alerts.py
+++ b/05_scheduling/hackernews_alerts.py
@@ -2,7 +2,7 @@
 # lambda-test: false
 # tags: ["featured"]
 # ---
-# # Hacker News Slackbot
+# # Run cron jobs in the cloud to search Hacker News
 
 # In this example, we use Modal to deploy a cron job that periodically queries Hacker News for
 # new posts matching a given search term, and posts the results to Slack.
diff --git a/06_gpu_and_ml/llm-serving/vllm_inference.py b/06_gpu_and_ml/llm-serving/vllm_inference.py
index 2023281ca..c3b4ebe08 100644
--- a/06_gpu_and_ml/llm-serving/vllm_inference.py
+++ b/06_gpu_and_ml/llm-serving/vllm_inference.py
@@ -5,7 +5,7 @@
 # tags: ["use-case-lm-inference", "featured"]
 # ---
 
-# # Run an OpenAI-Compatible vLLM Server
+# # Run OpenAI-compatible LLM inference with LLaMA 3.1-8B and vLLM
 
 # LLMs do more than just model language: they chat, they produce JSON and XML, they run code, and more.
 # This has complicated their interface far beyond "text-in, text-out".
diff --git a/06_gpu_and_ml/llm-structured/jsonformer_generate.py b/06_gpu_and_ml/llm-structured/jsonformer_generate.py
index 1295d5fa4..39ff324e2 100644
--- a/06_gpu_and_ml/llm-structured/jsonformer_generate.py
+++ b/06_gpu_and_ml/llm-structured/jsonformer_generate.py
@@ -1,7 +1,7 @@
 # ---
 # lambda-test: false
 # ---
-# # Generate synthetic data with Jsonformer
+# # Structured output generation with Jsonformer
 #
 # [Jsonformer](https://github.com/1rgs/jsonformer) is a tool that generates structured synthetic data using LLMs.
 # You provide a JSON spec and it generates a JSON object following the spec. It's a
diff --git a/06_gpu_and_ml/obj_detection_webcam/webcam.py b/06_gpu_and_ml/obj_detection_webcam/webcam.py
index e94c3f4ed..b2296570a 100644
--- a/06_gpu_and_ml/obj_detection_webcam/webcam.py
+++ b/06_gpu_and_ml/obj_detection_webcam/webcam.py
@@ -3,7 +3,7 @@
 # deploy: true
 # tags: ["use-case-image-video-3d", "featured"]
 # ---
-# # Machine learning model inference endpoint that uses the webcam
+# # Real-time object detection via webcam
 #
 # This example creates a web endpoint that uses a Huggingface model for object detection.
 #
diff --git a/06_gpu_and_ml/stable_diffusion/flux.py b/06_gpu_and_ml/stable_diffusion/flux.py
index 8c0dc4e92..649584c0a 100644
--- a/06_gpu_and_ml/stable_diffusion/flux.py
+++ b/06_gpu_and_ml/stable_diffusion/flux.py
@@ -4,7 +4,7 @@
 # tags: ["use-case-image-video-3d", "featured"]
 # ---
 
-# # Run Flux fast with `torch.compile` on Hopper GPUs
+# # Run Flux fast on H100s with `torch.compile`
 
 # In this guide, we'll run Flux as fast as possible on Modal using open source tools.
 # We'll use `torch.compile` and NVIDIA H100 GPUs.
diff --git a/06_gpu_and_ml/text-to-video/mochi.py b/06_gpu_and_ml/text-to-video/mochi.py
index bc0365371..a8dda2a56 100644
--- a/06_gpu_and_ml/text-to-video/mochi.py
+++ b/06_gpu_and_ml/text-to-video/mochi.py
@@ -2,7 +2,7 @@
 # cmd: ["modal", "run", "--detach", "06_gpu_and_ml/text-to-video/mochi.py", "--num-inference-steps", "64"]
 # ---
 
-# # Generate videos from text prompts with Mochi
+# # Text-to-video generation with Mochi
 
 # This example demonstrates how to run the [Mochi 1](https://github.com/genmoai/models)
 # video generation model by [Genmo](https://www.genmo.ai/) on Modal.
diff --git a/06_gpu_and_ml/yolo/finetune_yolo.py b/06_gpu_and_ml/yolo/finetune_yolo.py
index 985b9ada4..473d9d7bf 100644
--- a/06_gpu_and_ml/yolo/finetune_yolo.py
+++ b/06_gpu_and_ml/yolo/finetune_yolo.py
@@ -2,7 +2,7 @@
 # args: ["--no-quick-check"]
 # tags: ["use-case-image-video-3d", "use-case-finetuning"]
 # ---
-# # Fine-Tuning and Inference for Computer Vision with YOLO
+# # Fine-tune open source YOLO models for object detection
 #
 # Example by [@Erik-Dunteman](https://github.com/erik-dunteman) and [@AnirudhRahul](https://github.com/AnirudhRahul/).