Prompt Engineering Notes Prompt engineering is the process of designing high-quality prompts to guide LLMs to produce accurate outputs. This involves refining length, structure, tone, and clarity.
Model-Specific Setup Choose Your LLM and Configure: Model-specific prompts and capabilities Sampling parameters: Output Length
More tokens = more compute cost. Reducing length just truncates output. ReAct model warning: Emits irrelevant tokens post-response. Sampling Controls Parameter Effect Temperature Low = deterministic; High = creative/random Top-K Limit prediction to top K likely tokens Top-P Nucleus sampling = choose from top cumulative probability P Num Tokens Max output length Practical Guidelines: Temperature = 0.Read more...
but only when Marketing is done with intent… I’ve bought things. Hired people. Paid for services.
Not because they chased me. But because they told me who they were. What they do. What they offer.
I checked reviews. I saw their work. I decided based on what they made visible.
Imagine if they stayed silent. Too “humble” to speak. Too “pure” to sell.
I’d never have known. They’d never have helped.Read more...
Perfection is a Sham Delayed progress Analysis paralysis Decision fatigue Brain rot Sunk cost The elephant that forgot it could move Lost motivation Burnout Discontentment “Work expands so as to fill the time available for its completion.”
— Parkinson’s Law, C. Northcote Parkinson (1955)
And still, perfection is sought.
Craved. Chased. Worshipped.
Why?
Because perfection is a shield.
A shield against criticism.
A way to avoid failure.
A distraction from fear.Read more...
List of Links Slack Architecture G1GC Prompt Engineering White Paper Growth Map - Non Checklist 🔭 Concepts to Explore Later
Hybrid Search (Sparse + Dense Retrieval)
Combine traditional keyword search (like TF-IDF/BM25) with embeddings for better relevance, especially in enterprise search.
Vector Databases (FAISS, Pinecone, Weaviate, Milvus)
Each has trade-offs in latency, scalability, and integrations. Worth exploring for hands-on projects.
Document Chunking Strategies
How to split large docs into semantically meaningful chunks before embedding — affects RAG accuracy.Read more...
This wasn’t meant to be my first post. But I wanted to start with something real.
What’s happened to LinkedIn? Seriously, LinkedIn is overflowing with humble brags, pointless posts, and thinly veiled marketing. The good old “HR meets candidates” platform is now a full-blown social network. Everyone seems to have sudden epiphanies — some insightful, others… well, less so. Here’s one: learning about customer obsession from a street-side peanut vendor.Read more...