How I adapted the reactive agent pattern from academic review writing to create a personalized news curation system that fights information overload through intelligent deduplication and search engine orchestration.
From processed content to final PDF - document assembly, LaTeX compilation, performance optimization, and the caching strategies that make complex AI workflows practical.
Ensuring citation accuracy through automated bibliography processing - duplicate detection, format validation, and reference consistency in AI-generated academic content.
Deep dive into the sequential section writing process - from subsection generation to quality review loops, managing context and maintaining academic standards.
A detailed exploration of the reactive agent pattern - the core component that handles tool interactions, context management, and structured output extraction in complex AI workflows.
How the system generates table of contents through web research and plans word allocation across sections - the foundational structure that guides the entire writing process.
Deep dive into the multi-agent architecture using LangGraph - state management, workflow orchestration, and the divide-and-conquer approach to complex AI tasks.
Exploring the challenge of automatically generating comprehensive review articles from web research - defining the problem space and core requirements.
Learn how to implement intelligent related posts using local AI embeddings with Ollama, featuring semantic similarity, caching, and deployment strategies for Astro blogs.
The Related Posts feature pulls relevant content from your blog to display at the bottom of your posts. It uses your post content, tags, and categories to ...
Gatsby is a free and open source framework based on React that helps developers build blazing fast websites and apps. It combines the control and scalability of...
If you have apple airpods or airpods pro, and sometimes want to use it with your Ubuntu machines, follow this guide! If you try to pair airpods without doing any of these steps, most likely the pairing process will never complete.
Its been quite long since I last made any update to my Desktop at home. After a full year of working from home, I finally decided to make it more modern. My old desktop was a decent workhorse outlined here...
Balushahi is a traditional dessert originating from the Indian subcontinent. It is similar to a glazed doughnut in terms of ingredients, but differs in ...
A samosa or singara is a fried South Asian pastry with a savoury filling, including ingredients such as spiced potatoes, onions, and peas, or meat or fish.
YOLOv4 is a one-stage object detection model that improves on YOLOv3 with several bags of tricks and modules introduced in the literature. The components section below details the tricks and modules used.
Gatsby themes are plugins that include a gatsby-config.js file and add pre-configured functionality, data sourcing, and/or UI code to Gatsby sites. Novel theme...
The essential guide by one of America's leading doctors to how digital technology enables all of us to take charge of our health A trip to the doctor is almost a guarantee of misery. You'll make an appointment months in advance...
This guide covers the basics of setting up a development environment on a new Mac. Whether you are an experienced programmer or not, this guide is intended for...
fish is a Unix shell with a focus on interactivity and usability. Fish is designed to give the user features by default, rather than by configuration. Fish is considered an exotic shell since it does not rigorously adhere to POSIX shell standards, at the discretion of the maintainers.
Gatsby is an open-source static site generator built on top of Node.js using React and GraphQL. It provides large number of plugins to create static sites based on sources as Markdown, MDX, images etc.
First of all, apologies for a sudden update of the domain name - from datasciencevision.com to the new address of reckoning.dev! I let the old address lease expire and opted for a new (hopefully better!) home for this blog.
Comet ML is an experimentation platform, which permits testing Machine Learning projects, from the beginning up to the final monitoring. In this post, we would go through some examples to use comet.
An autoencoder is a type of artificial neural network used to learn efficient codings of unlabeled data. An autoencoder learns two functions: an encoding function that transforms the input data, and a decoding function that recreates the input data from the encoded representation.
GNOME, originally an acronym for GNU Network Object Model Environment, is a free and open-source desktop environment for Linux and other Unix-like operating systems.
Sublime Text is a shareware text and source code editor available for Windows, macOS, and Linux. It natively supports many programming languages and markup languages.
The Boosted Trees Model is a type of additive model that makes predictions by combining decisions from a sequence of base models. Gradient Boosted Machine (GBM) is one of the most effective machine learning models for predictive analytics, making it an industrial workhorse for machine learning.
Tree-based models use a decision tree to represent how different input variables can be used to predict a target value. Machine learning uses tree-based models for both classification and regression problems.
In the last post on SVM, we looked at the mathematical details of the algorithm. In this post, I will be discussing the practical implementations of SVM for classification as well as regression.
Nikola is a static site and blog generator based on Python. However, Hugo is one of the most popular open-source static site generators. With its amazing speed and flexibility, Hugo makes building websites fun again!
A descriptive statistic is a summary statistic that quantitatively describes or summarizes features from a collection of information, while descriptive statistics is the process of using and analysing those statistics.
In machine learning, support vector machines are supervised learning models with associated learning algorithms that analyze data for classification and regression analysis. In this post, we will go over mathematical details of SVM models.
Gajar ka halwa, also known as gajorer halua, gajrela, gajar pak, and carrot halwa is a carrot-based sweet dessert pudding made by placing grated carrots in a pot containing a specific amount of water, milk and sugar, cardamom and then cooking while stirring regularly.
The world of computers is moving fast. While going through some materials on algorithms, I have come across an interesting discussion -enhancements in hardware (cpu) vis-a-vis algorithms. One side of the coin is the hardware - the speed of computers, while the other is of algorithms...
KDE Plasma 5 is the fifth and current generation of the graphical workspaces environment created by KDE primarily for Linux systems. In this guide, we will go over its installation in Arch Linux.
Arch Linux is an independently developed, x86-64 general-purpose Linux distribution that strives to provide the latest stable versions of most software by following a rolling-release model. The default installation is a minimal base system, configured by the user to only add what is purposely required.
Two simple maths puzzles! In the first problem we prove a well known property of prime numbers. In the second one, we are just trying to get shortest path on a chess board.
Palak/Spinach paneer is a common north indian cuisine, indian cottage cheese cooked in spinach puree. Its a bit involved than my last few dishes. Nevertheless, I promise you - the yummyness of this one is worth all the trouble!
Typically, hunger and laziness come to me as inseparable couples. To make things worse, I have been trying to eat healthy. Nevertheless, here is another recipe that solves all these at once.