diff --git a/package-lock.json b/package-lock.json
index 1611a389..f2f45e25 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -1,5 +1,5 @@
 {
-  "name": "blacksheepcode",
+  "name": "blacksheepcode2",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
diff --git a/src/assets/yak_shaving.jpeg b/src/assets/yak_shaving.jpeg
new file mode 100644
index 00000000..8f16b401
Binary files /dev/null and b/src/assets/yak_shaving.jpeg differ
diff --git a/src/routes/drafts/psychological_pitfalls_of_ai_urgency.mdx b/src/routes/drafts/psychological_pitfalls_of_ai_urgency.mdx
new file mode 100644
index 00000000..3e65bacf
--- /dev/null
+++ b/src/routes/drafts/psychological_pitfalls_of_ai_urgency.mdx
@@ -0,0 +1,15 @@
+---
+meta:
+  title: "Psychological pitfalls of using artificial intelligence - a heightened sense of urgency"
+  description: Because AI can so quickly provide 
+  dateCreated: 2026-05-06
+
+tags: 
+  - "ai"
+  - "software_engineering"
+---
+
+Something that I've found happening as I've started using AI tools more, especially agentic AI workflows, is that the AI very quickly comes back with a solution, and it puts me in this frenetic state where it feels like I need to make a decision on this thing now. 
+
+
+There's 
\ No newline at end of file
diff --git a/src/routes/posts/agentic_ai_yak_shaving_on_steroids.mdx b/src/routes/posts/agentic_ai_yak_shaving_on_steroids.mdx
new file mode 100644
index 00000000..48756468
--- /dev/null
+++ b/src/routes/posts/agentic_ai_yak_shaving_on_steroids.mdx
@@ -0,0 +1,36 @@
+---
+meta:
+  title: Agentic AI - yak shaving on steroids
+  description: The prospect of having agentic AI do all the real work means that we now focus on tuning the AI, not delivering the work.
+  dateCreated: 2026-05-10
+  image: "yak_shaving"
+
+tags: 
+  - "ai"
+  
+---
+import yak_shaving from "@/assets/yak_shaving.jpeg";
+
+I've been playing around with agnetic AI for a side project recently, and I can absolutely see a world where the workflow is like this: 
+
+1. I have an iterative session with an AI to define a feature - the outcome of this is an AI generated brief
+2. The AI goes off and implements the feature, unsupervised, the outcome of this is a pull request. 
+3. I review the pull request - what I'm looking for is a cursory look over of the code to check that the patterns are holding, and brief manual verification that the feature is working. 
+
+_This is not my current workflow_ - that step 2. I have to supervise - because the AI goes off into rabbit holes and burns all my tokens and creates some messes. 
+
+But this worfklow seems tantalisingly in reach - whether that's all a fugazi is a different story. 
+
+I have limited time. I work a full time job, and I have a four-year old and two four-month old babies, and I need to exercise. So with my side project, of which I'm about 40 hours in, in, I could spend my limited time chipping away at building new features, or I could spend it: 
+
+- Optimising my prompts
+- Adding lint rules and other guard rail test scripts
+- Adding generation scripts
+
+In in the hope that at the end of that the end of it, I'll be able to to just informally have a chat session with an AI, creating a list of features I want to see, and then come back the next night to review and merge the code it has created. 
+
+<img src ={yak_shaving.src} alt="ai generated image. A man is scoffing some pills while holding some shears. In the background are some yaks. The man is saying 'gonna go shave me some yaks'" className="blog-image" style={{height: 400, display: "block", margin: "40px auto"}}/>
+
+
+This is feels like yak-shaving. Perhaps this is another bit of software engineering wisdom that is disrupted by AI. 
+