Daily Dose of Data Science
Subscribe
Sign in
Home
Sponsor
Premium
Archive
Leaderboard
About
Latest
Top
Discussions
Finding Every Vulnerability Made Security Harder
+ building an agent engineering team with Grok Bot!
11 hrs ago
•
Avi Chawla
5
Grok Bot Masterclass
...covered with a full hands-on video demo.
Nov 4
•
Avi Chawla
6
1
Agent Framework vs. Agent Harness, clearly explained!
+ a technique for 23x faster attention!
Nov 3
•
Avi Chawla
12
2
12 Optimizers for Training Neural Networks, clearly explained!
...and what each one costs in GPU memory!
Nov 2
•
Avi Chawla
3
2
October 2026
Why Other Requests Change Your LLM’s Output
+ an update on Harness Engineering 2.0.
Oct 31
•
Avi Chawla
11
3
RAG vs. Vectorless RAG, clearly explained!
...explained visually (with tradeoffs in production).
Oct 30
•
Avi Chawla
7
2
Prefix-Aware LLM Routing, clearly explained
...explained visually (the trick to get the first LLM token 170x sooner).
Oct 29
•
Avi Chawla
8
1
Why DeepSeek stores its KV cache on disk
DeepSeek's engineering behind 26x faster time to first token.
Oct 28
•
Avi Chawla
8
1
Which OCR Solution Works Best for Your Documents?
...DeepSeek-OCR vs. PaddleOCR-VL vs. Datalab Surya, tested on the same 150 pages.
Oct 27
•
Avi Chawla
7
1
1
Why Temperature 0 Isn't Deterministic in LLM Inference
...explained visually (a popular LLM interview question).
Oct 26
•
Avi Chawla
10
4
How CUDA Graph Replay Cuts Inference Overhead
...explained visually.
Oct 24
•
Avi Chawla
4
2
1
6 KV Cache Quantization Approaches
...explained visually.
Oct 23
•
Avi Chawla
6
2
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
