@@ -56,7 +56,8 @@
 
 APPLICATION_NOTICE = (
     "4 weeks over 5 calendar weeks (~8-10 h/week), free and open-source. "
-    "Welcome 10 Nov 2026 · starts 17 Nov 2026 · apply by 26 Nov 2026 · ends 21 Dec 2026 · capstone due 15 Jan 2027. "
+    "Welcome 10 Nov 2026 · starts 17 Nov 2026 · apply by 26 Nov 2026 · "
+    "ends 21 Dec 2026 · capstone due 15 Jan 2027. "
     "Prerequisites: Python, Docker basics, HTTP APIs, command line. "
     "Only applicants whose background fits the medium-advanced level "
     "are approved."
@@ -65,11 +66,14 @@
 MARKETING_MARKDOWN = """\
 ## Production AI Agents for Generative Media
 
-4 weeks over 5 calendar weeks (Module 1 is two weeks) · ~8-10 h/week · free · open-source · project-based.
+4 weeks over 5 calendar weeks (Module 1 is two weeks) · ~8-10 h/week ·
+free · open-source · project-based.
 
-> How do you build a production AI system when an agent can launch expensive, slow, failure-prone GPU workloads that produce large media artifacts?
+> How do you build a production AI system when an agent can launch expensive,
+> slow, failure-prone GPU workloads that produce large media artifacts?
 
-Outcome: build, deploy, evaluate, optimize and monitor a production AI system in which agents orchestrate GPU-powered image, video and audio models on Nebius Serverless AI.
+Outcome: build, deploy, evaluate, optimize and monitor a production AI system
+in which agents orchestrate GPU-powered image, video and audio models on Nebius Serverless AI.
 
 ### Dates (2026 cohort)
 
@@ -85,8 +89,11 @@
 
 ### Program
 
-- **Week 1 - Make it work.** Generative media workloads: modalities and model families, first GPU generation from script to container to GPU job, containers for generative AI, GPU engineering, jobs vs endpoints.
-- **Week 2 - Make it intelligent.** Agents that use GPUs: the agent loop, tools and structured outputs, CPU agents with GPU tools, long-running tools.
+- **Week 1 - Make it work.** Generative media workloads: modalities and model
+  families, first GPU generation from script to container to GPU job,
+  containers for generative AI, GPU engineering, jobs vs endpoints.
+- **Week 2 - Make it intelligent.** Agents that use GPUs: the agent loop,
+  tools and structured outputs, CPU agents with GPU tools, long-running tools.
 - **Week 3 - Make it scale.** Scaling, reliability and cost.
 - **Week 4 - Make it production-ready.** Evaluation, observability and production.
 
@@ -98,14 +105,17 @@
 
 ### Prerequisites
 
-Python, Docker basics, HTTP APIs, command line. No diffusion theory required. Prior LLM or MLOps course experience helps, not required.
+Python, Docker basics, HTTP APIs, command line. No diffusion theory required.
+Prior LLM or MLOps course experience helps, not required.
 
 Application only, limited seats.
 """
 
 META_DESCRIPTION = (
-    "Production AI Agents for Generative Media - free project-based course on production GPU media systems. "
-    "Welcome 10 Nov 2026, starts 17 Nov, apply by 26 Nov, ends 21 Dec, capstone 15 Jan 2027. Limited seats, application only."
+    "Production AI Agents for Generative Media - free project-based "
+    "course on production GPU media systems. "
+    "Welcome 10 Nov 2026, starts 17 Nov, apply by 26 Nov, ends 21 Dec, "
+    "capstone 15 Jan 2027. Limited seats, application only."
 )