stephenleo

The Definitive Guide to Structured Data Parsing with OpenAI GPT 3.5 in Complex Problems

Code for the Medium post

stephenleo

Semantic Caching in Generative AI Chatbots

Code for the Medium post

stephenleo

Big Query and SQL Hacks

1. Use QUALIFY to filter rows after a window function. It's way more computationally efficient than using a CTE.

   SELECT RANK() OVER (PARTITION BY product_id ORDER BY rating DESC) AS rank 
   FROM table
   QUALIFY rank = 1
   

stephenleo

ArXiv Semantic Paper Search

• Code from the Medium post Link

stephenleo

Semantic Textual Similarity

Code for the Medium post Link

stephenleo

DBS (Development Bank of Singapore) Data Analyst Interview

1. Python puzzle
   • Given a $2 \times 1$ non-negative integer array $\vec{a}$ and a $1 \times n$ integer array $\vec{b}$, write a function to return a $2 \times n$ logical matrix $x_{ij}$ such that $\sum_{j=1}^{n}x_{ij} = a_i$ and $\sum_{i=1}^{2}x_{ij} = b_j$. If there is no working matrix, please return string "impossible" (15 mins).
   • What is logical matrix? This is a matrix that $x_{ij} \in {0,1}$
   • Example: $\vec{a} = \begin{bmatrix} 3 \ 2\end{bmatrix}$, $\vec{b} = \begin{bmatrix} 2 & 1 & 1 & 0 & 1\end{bmatrix}$, one possible matrix is $\begin{bmatrix} 1 & 1 & 1 & 0 & 0 \ 1 & 0 & 0 & 0 & 1\end{bmatrix}$
   • Solution:

     import numpy as np
     

stephenleo

Make Money on the side with Data Science!

Code for the Medium post Link

stephenleo

Supercharged Semantic Similarity Search in Production

Code for the Medium post Link

stephenleo

Python

A collection of useful Python snippets

stephenleo

Useful ES commands

1. Check available nodes: curl 'http://<ip_address>:9200/_cat/nodes?v'
2. Check available indices: curl 'http://<ip_address>:9200/_cat/indices?v'
3. Delete a specific index: curl -X DELETE 'http://:9200/?pretty'