How to convert documents to Markdown with MarkItDown

Built a quick reference guide for RAG patterns. Spent some time documenting 5 common patterns I kept seeing in production systems: • Semantic chunking • HyDE (query expansion) • Re-ranking • Metadata filtering • Query decomposition Each one has working Python code and notes on when it's worth using vs. when it's overkill. Also threw in some case studies with actual numbers ($900K-$2.3M impact range) from real implementations I researched. Nothing groundbreaking - just a clean reference for the trade-offs (latency, cost, quality) since I couldn't find one that laid it out clearly. Live docs: https://lnkd.in/edsUB5mA Code: https://lnkd.in/e9Xqrme6 #MachineLearning #RAG #Python

🧩 Day 32 — Find the Difference of Two Arrays (LeetCode 2215) 📝 Problem Given two 0-indexed integer arrays nums1 and nums2, return a list answer of size 2 where: - answer[0] contains all distinct integers in nums1 not present in nums2. - answer[1] contains all distinct integers in nums2 not present in nums1. The order of elements in the output lists does not matter. 🔁 Approach - Use sets to eliminate duplicates and enable fast lookups. - Convert both arrays to sets: set1 and set2. - Compute differences using set operations: - set1 - set2 gives elements in nums1 not in nums2. - set2 - set1 gives elements in nums2 not in nums1. - Convert the results back to lists to match the required output format. 📊 Complexity - Time Complexity: O(n + m) - Space Complexity: O(n + m) 🔑 Concepts Practiced - Set operations for fast difference computation - Removing duplicates using set() - Efficient list transformation - Clean and readable Python logic #Python #DSA #Leetcode #ProblemSolving #Arrays

Hey Everyone ! I have Just Finished a simple #python Script To understand huffman's algorithm In case Anyone Interested In learning Read this Below * Huffman’s algorithm is a way to compress data (make it smaller) by using shorter codes for frequent characters and longer codes for rare ones. And The steps Are : > Count frequency of each character in the data. > Build a tree > Assign binary codes > Encode the data Wanna Check the Code? Here is my github repo : https://lnkd.in/dAi7cqZf Have a Great Day Guys ! Any questions Feel free to ask me. hashtag #Algorithms hashtag #DataStructures hashtag #ComputerScience

 Visualizing Data with Box Plots in Python Today, I explored one of my favorite tools for data exploration the box plot! Box plots are powerful for spotting outliers, understanding data distribution, and comparing feature ranges at a glance. In this example, I visualized multiple features from a dataset using: df[var].plot(kind='box', figsize=(20, 4), subplots=True) With just one line of code, I got a clear picture of how each variable behaves from acidity levels to alcohol content. Key insight: Outliers tell a story. Instead of rushing to remove them, I always pause to ask why they exist. Sometimes, they reveal patterns worth exploring. Have you used box plots in your EDA before? What’s your go-to visualization for spotting outliers? #DataAnalysis #Python #EDA #DataVisualization #BoxPlot #Pandas #DataScience

Excel is great for quick analysis, but it becomes less effective when your data gets bigger or your formulas become more complex. That’s where Python in Excel comes in. It lets you run Python code right inside your spreadsheet — no switching tools, no manual workarounds. In this DataCamp article, I explore how to use Python in Excel for advanced analytics, visualizations, and even machine learning, all within your familiar workflow. Read it here: https://lnkd.in/dHWFVFjB #python #excel #analytics

Problem: Raw supercapacitor test files from the lab are massive and hard to analyze. Solution: I built a custom Python app to handle it. My tool uses Pandas to process all 175+ cycles, then Matplotlib to visualize the results in a Tkinter GUI. It automatically generates key diagnostic plots, including degradation curves, V-Q profiles, and a full Energy vs. Power Ragone plot. The final feature? A one-click "Export to PDF" button that packages the entire analysis into a professional report. #Python #Automation #DataAnalysis #Engineering #Supercapacitor #R&D #DataVisualization #Pandas #Matplotlib

🔁 Day 45 of #100DaysOfDSA — Reverse Linked List Problem: Reverse Linked List (LeetCode #206) Given the head of a singly linked list, reverse it and return the reversed list. Concepts Used: 🧠 Stack-based approach Traverse the linked list and push each node’s value into a stack. Then reassign values by popping from the stack to reverse the list. 🧩 Learning Reflection: This problem helped me understand how stacks can simplify linked list manipulations. Though not the most space-efficient solution, it reinforces the concept of LIFO (Last In, First Out) operations beautifully. ⏱️ Complexity: Time: O(N) Space: O(N) 💬 Closing Thought: Sometimes, using an extra data structure makes logic clearer — and that’s okay while learning. #LeetCode #DSA #LinkedList #Python #100DaysOfCode #CodingJourney

Outliers may look like tiny dots in your data, but they can drastically impact your analysis, KPIs, and model performance. In this presentation, I’ve shared how to: 1️⃣ Detect outliers using boxplots 2️⃣ Calculate upper and lower limits using IQR 3️⃣ Handle outliers effectively using capping Small change in preprocessing can make a big difference in insights and model accuracy. #DataScience #MachineLearning #Outliers #Python #IQR #EDA #DataPreprocessing #Analytics #uptor

#96day of #100DaysOfCode 🌱 LeetCode 109: Convert Sorted List to Binary Search Tree Today’s challenge was about building balance — literally! Given a sorted linked list, we need to convert it into a height-balanced BST 🌳 🧠 Key Idea: The middle element of the list becomes the root. Left half forms the left subtree, right half forms the right subtree. Use slow–fast pointers to find the middle efficiently. 📈 Complexity: Time: O(n log n) Space: O(log n) (recursion stack) 💬 Learning: Balance matters — in trees, in code, and in life 🌿 Sometimes, the middle point creates the strongest foundation. #LeetCode #DSA #BinarySearchTree #LinkedList #CodingChallenge #Python #Algorithm #LeetCodeDaily #100DaysOfCode

Day 9 of #100DaysOfLeetCode Problem: 21. Merge Two Sorted Lists Category: Linked List / Two Pointers Today’s challenge focused on merging two sorted linked lists into a single sorted list. This problem was a great refresher on pointer management, conditional linking, and efficient traversal through nodes. 🧠 Key Learnings: Used two pointers to compare nodes from both lists and attach the smaller node to the merged list. Managed the traversal smoothly without losing reference to the new head node. Reinforced the concept of dummy nodes for cleaner list initialization. Strengthened confidence in working with linked data structures and node connections. 🎯 Takeaway: Linked lists teach the importance of precise pointer handling — one wrong reference can change the entire structure! #LeetCode #100DaysOfCode #ProblemSolving #CodingJourney #LinkedList #Pointers #Python #AIEngineer #Consistency