WTW Insurance Consulting and Technology’s Post

Step into the future with Radar Live Python, where open source meets streamlined deployment! Mehul Dave, P&C Practice Leader, APAC at WTW, unveils Radar Live Python, an innovative tool that revolutionizes insurance model deployment by integrating Python and open-source capabilities. This powerful extension reduces costs, simplifies control, and consolidates deployments into a single repository, enabling insurers to focus on innovation and value-added activities: https://ow.ly/B1bl50UiNlN #InsuranceTech #Innovation #WTW #Radar #Insurance

Today in the #C3S General Assembly: technical aspects and a forward look towards ERA6 and CMIP7, combined with the (wide) user uptake of current ERA5 products. The user focused strategy of C3S does pay dividend. Although the original API is interfaced through python (cdsapi), I hope the 50K users I serve through the BlueGreen Labs {ecmwfr} #rstats package.

📣 SigTech Framework v9 is Now Live! 📣 We're thrilled to announce the launch of Framework v9! Powered by Python 3.11, offering significantly faster speeds for your research environment via our Enterprise solution. Key Highlights: · Speed Enhancements: Python 3.11 ensures quicker strategy execution. · Advanced Reporting: New features and statistics for more detailed performance insights. · Better User Experience: Quality-of-life improvements with upgraded libraries, new default values, and refined framework logic. Explore the full range of enhancements and see how Framework v9 can elevate your strategy development. 🚀 🔗 Read the blog here: https://lnkd.in/ej47JMnX #SigTech #QuantitativeResearch #Python311 #FinTech #FrameworkUpdate

This video shows how to use Alpaca's API to download historical stock data, focusing on the S&P 500. It covers setting up a Python client, retrieving stock data via the stock bars function, and handling common issues like date formatting. The tutorial also touches on requesting crypto data and potential API errors.

For efficient Multi-Asset Portfolio management in QuantJourney Framework, we've added #PortfolioData and #InstrumentData classes and implemented advanced techniques to enhance portfolio optimization and our proprietary backtesting framework. It's fully vectorized, using #Numpy and #Numba (with @njit decorator for nopython mode compilation, to skip python GIL/interpreter if possible). This allows efficient processing of large datasets at high speed. Our #backtester now includes also Conditional Portfolio Optimization (#CPO) methods, enabling dynamic weight adjustments based on market conditions and custom events, so it's purely event-based backtester (3 types of #Events, not only typical signals). These enhancements have resulted in a 10-15x speed improvement for our largest multi-asset portfolios, enabling more comprehensive strategy testing and risk analysis. Read more at https://lnkd.in/dnE_qcy6  And get CODE in Python as subscriber.

2275: Largest Combination with bitwise and greater than zero The bitwise AND of an array nums is the bitwise AND of all integers in nums. For example, for nums = [1, 5, 3], the bitwise AND is equal to 1 & 5 & 3 = 1. Also, for nums = [7], the bitwise AND is 7. You are given an array of positive integers candidates. Evaluate the bitwise AND of every combination of numbers of candidates. Each number in candidates may only be used once in each combination. Return the size of the largest combination of candidates with a bitwise AND greater than 0. LeetCode Approach: - Creating a hash map which will store the bits addition. Indexing of the hash map will be in accordance with 2^(i) as the 0th index will store the addition of 2^0 bit values. - adding up all the bits in the element and returning the max value of the hash map. bin() function in python return the binary of the integer e.g. bin(10)="0b1010" so, taking the [2:-1] will return the bin value excluding the "0b"of the bin representation. #100daysofcode #100daysofchallenge #100daysofcoding #leetcode #challenge #day6

DSA:Problem No 9 #Geekstreak2024 We initialize two variables: max_height to keep track of the maximum height seen so far and count to count the number of buildings that can see the sunrise. We iterate over the heights of buildings from left to right. For each building, we check if its height is greater than the maximum height seen so far. If it is, we increment the count and update the maximum height seen so far. Finally, we return the count of buildings that can see the sunrise. This solution has a time complexity of O(n), where n is the number of buildings, and an auxiliary space complexity of O(1), as we only use a constant amount of space to store the maximum height and count. #Coding #Python #Geekstreak2024 #DeutscheBank

PyKX 3.0 brings groundbreaking updates to streamline workflows, including 4x faster conversions from Pythonic data types to PyKX 🚀 Watch our VP of Data Science Conor McCarthy break down the three core changes that make this speed possible. For a full look at all the PyKX 3.0 updates, check out our latest blog: https://lnkd.in/e_MEp8S5 #SystemUpdates #KX #PyKX #PyKX3.0 #Python #WorkflowStreaming #DataStreaming #PythonSyntax

Recently, while 𝐚𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐧𝐠 a 𝐁𝐚𝐜𝐤𝐄𝐧𝐝 workflow, I encountered a challenge. The task involved submitting an API request to initiate a process and later polling another API to check for its completion status. The goal was to ensure the process completed successfully by verifying a specific value (progress = 1) in the response. Initially, I implemented a fixed wait time between the API request and the status check once. While this approach worked well for smaller datasets, it started failing for larger datasets because the process took longer to complete. The fixed wait time was not adaptable to varying tenant sizes and processing times. To solve this, I switched to a retry mechanism. Here’s how it worked: - After submitting the initial API request, I implemented retries with an N-second interval (where N can be any value that suits the process). - The system kept polling for the completion status until the expected value was received or the retry limit was reached. - This approach allowed the workflow to account for varying processing times, reducing unnecessary waits. The retry mechanism eliminated the dependency on static wait times and made the workflow dynamic. #testautomation #backendautomation #pytest #python #retrymechanism