Every row is a frame. One frame shows a moment. Trends show the story. This is where real insights live. In questions like:  • Did revenue grow or stall?  • Is the change accelerating or slowing?  • Is today unusual or just noise? Some of the useful functions for this in Python are:  • 𝘀𝗵𝗶𝗳𝘁() adds context  • 𝗱𝗶𝗳𝗳() see the change compared with before  • 𝗽𝗰𝘁_𝗰𝗵𝗮𝗻𝗴𝗲() sometimes is easier to see the change as a relative number  • 𝗿𝗼𝗹𝗹𝗶𝗻𝗴() reveals local trends and smooths noise Thinking in differences forces better questions like:  • Compared to when?  • Is this change normal or surprising?

Computational Models in Finance: Stock Market Clustering 📈 I just completed a project using Unsupervised Machine Learning to optimize portfolio selection. By applying K-Means Clustering across multiple time horizons (2-week to 12-month), I was able to segment stocks by their risk-return profiles rather than just their industry. The Result? A data-driven recommendation engine that identifies high-potential assets—specifically optimized for a $1,500 budget. 🔹 Tools used: Python, Scikit-Learn, Pandas, Matplotlib. 🔹 Key Logic: Elbow Method for optimal 'k', volatility filtering, and budget-constrained scoring. Check out the full code and analysis on my GitHub: [https://lnkd.in/gY74QrJC] #FinancialEngineering #DataScience #Python #MachineLearning #StockMarket #InvestmentAnalysis

Mojo in the Age of AI Agents Wes McKinney's recent post on "agent ergonomics" hit a nerve—especially his take on why Python struggles when AI agents become primary code authors. His thesis: agents need fast compile-test cycles, standalone binaries, and predictable performance. Python's dynamic nature (slow feedback, dependency hell, runtime surprises) works against reliable agent iteration. Go and Rust shine here, but come with a learning curve for Python-heavy teams. Enter Mojo from Modular (Chris Lattner and Tim Davis) —the missing piece I've been exploring in my GPU programming work. Mojo is a Python superset designed specifically for AI/systems programming. It keeps Python's syntax that agents already know, but adds: ✅ Static typing + borrow checking (compile-time safety)  ✅ Sub-second compiles with no GIL or JIT unpredictability  ✅ Direct GPU/TPU control (10x+ Python speedups)  ✅ Seamless NumPy/Python interop (no ecosystem lock-in) The result? Agents can prototype in Python, then "mojofy" for production performance without retraining on Go's channels or Rust's ownership. I've written up how Mojo fits into Wes's landscape, with a side-by-side comparison to Go/Rust and a practical agent workflow example: 👉 https://lnkd.in/gKzzB8Va If you're building agent-driven data pipelines or AI workflows, Mojo might be the bridge to scalable code without a full rewrite. Curious what others think—are we really shifting from "human ergonomics" to "agent ergonomics" in language design? #Mojo #AIAgents #Python #PerformanceEngineering #DataScience

Stop solving for X by hand. 🛑 If you are still grinding through symbolic algebra on a notepad, you’re missing out on one of Python's most underrated superpowers: SymPy. Most developers think of Python for data manipulation (Pandas) or ML (Scikit-Learn). But as a mathematician, I’ve found that SymPy is the essential "missing link." Unlike NumPy, which handles numerical approximations, SymPy is a Computer Algebra System (CAS). It keeps your math "pure." Why should you care? - Symbolic Integrity: It doesn't turn 1/3 into 0.33333333. It keeps it as 1/3 until the very end. - Automated Solvers: As shown in the snippet below, you can define a complex polynomial and find its roots (even complex ones) in two lines of code. - LaTeX Ready: It typesets your results beautifully, making it a dream for documentation and research. The Pro-Tip: If you ever find SymPy hitting a performance wall, that’s your signal to look into SAGE. It’s the "big brother" that consolidates over 70 open-source packages for heavy-duty computation. Python isn't just for "coding"—it’s for extending the limits of how we solve problems. Are you Team SymPy for quick scripts, or have you made the jump to SAGE for full-scale mathematical modeling? #PythonProgramming #DataScience #Mathematics #MachineLearning #SymPy #CodingLife #STEM #ArtificialIntelligence #SoftwareEngineering

02 #AI_ML_for_Process_Engineering TRADING SPREADSHEETS FOR PYTHON WHY THE SWITCH FROM EXCEL? 🚀 SCALABILITY: Python handles millions of sensor readings that would crash a standard spreadsheet. 🛠️ REPRODUCIBILITY: Unlike Excel, where one accidental keystroke can break a formula across 10,000 rows, Python logic is explicit, modular, and verifiable. 📊 AUTOMATED INSIGHT: With one line of code (.describe()), I can instantly get the mean, std dev, and ranges for every tag in a massive dataset. Considering The 80/20 rule is real ==> 80% of AI is "Data Cleaning. Python is the power tool that makes that 80% manageable, allowing us to stop "firefighting" data and start interrogating it for insights. [Question for the Engineers] What is the largest dataset you’ve ever tried to open in a spreadsheet? Did it survive, or did you see the "Not Responding" screen of death? 😅 #DJ2Tech #ProcessEngineering #Industry40 #DigitalTransformation

AI Pathfinder – Visualizing Uninformed Search Algorithms I built a fully interactive search visualizer implementing six fundamental Uninformed Search strategies: • Breadth-First Search (BFS) • Depth-First Search (DFS) • Uniform-Cost Search (UCS) • Depth-Limited Search (DLS) • Iterative Deepening DFS (IDDFS) • Bidirectional Search The application animates each algorithm step-by-step, clearly showing: Expansion order of nodes Frontier vs explored states Final optimal path Performance comparison in best and worst-case grid scenarios This project strengthened my understanding of search theory, branching behavior, completeness, optimality, and time/space trade-offs — beyond just textbook concepts. Built using Python, NumPy, and Matplotlib with a fully interactive GUI. GitHub Repository: https://lnkd.in/d3vU2tch #ArtificialIntelligence #SearchAlgorithms #Python #ComputerScience #AIEngineering #MachineLearningJourney

Hello, guys!! For the past few days I have been working on the Black-Scholes options pricing model, where I tried to implement some basic machine learning concepts into the model, where I used random forest to predict the volatility. While building this project, I have used many Python packages like NumPy, SciPy.stats, yfinance, Matplotlib, Pandas, and Sklearn. - What the project does - Pulls live stock prices, option expiries, and option chains using yfinance - Automatically selects the ATM strike - Prices European call & put options using the Black–Scholes model - Computes key Greeks: Delta, Gamma, Vega, Theta - Estimates historical volatility from log returns - Trains a Random Forest model to predict short-term future volatility - Reprices options using ML-predicted volatility - Computes implied volatility from market prices using numerical methods - Compares market prices vs theoretical prices - Visualises price, delta, and gamma curves #QuantFinance #OptionsPricing #BlackScholes #MachineLearning #Python #Derivatives #FinancialEngineering #DataScience #Volatility #Greeks

Just wrapped up the Longest Consecutive Sequence problem on LeetCode and wrote a short breakdown on Medium. It’s a fun example of how a problem that looks like it needs sorting can actually be solved in O(n) time using a set and a simple insight about sequence starts. Small optimization, big payoff. This also marks the end of the Arrays & Hashing section of NeetCode 150 for me. Two pointers are up next, and for the first time in a while, burnout feels like it’s finally backing off. If you’re learning DSA or prepping for interviews, this one is worth a look. Read more here: https://lnkd.in/gWZ4WQmP #LeetCode #DataStructures #Algorithms #Python

📊 Multiple Linear Regression — Visualized in 3D I implemented Multiple Linear Regression completely from scratch using NumPy and visualized how a regression plane fits the data when we have two input features. 🔄 The GIF shows the regression plane rotating in 3D to build intuition about: • how multiple features influence predictions • why linear models form planes (not lines) • why visualization breaks beyond 2 features After visualization, I evaluated the same implementation on higher-dimensional datasets using R² score, since plots are no longer possible. 🔗 Full implementation & clean repo: https://lnkd.in/gDzhWKAW Learning ML by building from scratch 🚀 #MachineLearning #MultipleLinearRegression #MLFromScratch #Python #NumPy #DataScience