Jonathan Pedoeem

AI Agents in Military Use: integrating LLMs and Autonomous Systems on the Frontline

When Vladimir Putin declared in 2017 that "whoever leads in AI will dominate the world," he predicted future wars would be fought by autonomous drones rather than human soldiers. That prediction has become reality with chilling speed. In Ukraine today, drones are reported to cause 70-80% of

Bringing the Fundamentals to AI Engineering

AI engineering is a new discipline, but that doesn't mean we should throw out everything we know about engineering. The same fundamentals apply: de-scope ruthlessly, think in functions, and don't build what you don't need. Too many are skipping the fundamentals. Marketing Outpaced

How to Fine-Tune a Translation Model

Here's a surprising fact: you can unlock a large language model's translation abilities with as few as 32 parallel examples. This discovery is revolutionizing how we approach specialized translation, making high-quality translation accessible for medical journals, legal documents, literary works, and even endangered languages, all

AI Chatbot Conversations Archive

The LMSYS-Chat-1M dataset exposed the sheer scale of modern chatbot interactions, 1 million real-world conversations from 210,000 unique users with state-of-the-art language models. This revelation is a window into the massive infrastructure quietly recording every exchange between humans and AI systems worldwide. Behind

Grok 5: What We Expect

Elon Musk recently made a striking claim that drew attention across the AI community: Grok 5 has a "10% and rising" chance of achieving AGI—or something indistinguishable from it. This is more than a routine tech-CEO prediction; it is a direct challenge to the AI establishment

GPT-5 Context Window: The AI Memory Revolution That Changes Everything

Like Moore's Law transformed computing power, context windows in AI have been doubling every year, and GPT-5's massive 400K-token leap represents the most dramatic expansion yet. This isn't just another incremental upgrade; it's a fundamental shift in what artificial intelligence

Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights Synthesis

What if you could customize an AI model in seconds, just by describing what you want it to do? This is the revolutionary promise of Drag-and-Drop LLMs, a breakthrough approach that transforms how we adapt language models for specific tasks. Traditional model fine-tuning through methods like LoRA

Composer: What Cursor's New Coding Model Means for LLMs

Cursor just released Composer, an AI model that completes coding tasks in under 30 seconds, 4× faster than comparable systems, and it's trained inside real codebases using reinforcement learning. This isn't merely an incremental improvement to existing AI coding assistants; Composer represents a fundamental shift from

text-embedding-3-small: High-Quality Embeddings at Scale

OpenAI made embeddings both stronger and 5× cheaper: text-embedding-3-small improves on its predecessor while costing just $0.02 per million tokens. The model transforms text into 1536-dimensional vectors for semantic search, clustering, and RAG applications, delivering a major cost-efficiency gain for production AI systems and

AI Agents in Military Use: integrating LLMs and Autonomous Systems on the Frontline

Bringing the Fundamentals to AI Engineering

How to Fine-Tune a Translation Model

AI Chatbot Conversations Archive

Grok 5: What We Expect

GPT-5 Context Window: The AI Memory Revolution That Changes Everything

Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights Synthesis

Composer: What Cursor's New Coding Model Means for LLMs

text-embedding-3-small: High-Quality Embeddings at Scale

The first platform built for prompt engineering

Usage

Company

Follow Us