Practical AI applications
for real-world impact
We're an independent studio of passionate engineers and researchers, driven to explore and advance the frontier of AI through open research and real-world software applications.
Practical AI Applications that
Drive Innovation
Research & Open Source
We run small, focused research projects to explore practical questions in AI. We share our results in white papers and open-source tools — contributing to the community while deepening our own understanding.
Explore our ResearchAI Consulting & Development
We help companies design and build AI-enabled software — from prototypes to production. Our work focuses on LLMs, context-aware assistants, and internal tools that create real business value.
Work with usWe're open to contributors on our research side and available for select client work.
Research & Insights
We share our findings, experiments, and learnings with the community through whitepapers, research articles, and blog posts.
AI Eats Software.Trading Determinism for Probability
Most business apps are CRUD wrapped in business logic, and that logic is migrating into an AI agent layer. "SaaS is dead" sounds right until you ask where reliability actually comes from. Capability was never the line. Verifiability is. This essay maps what AI will eat, what it won't, and why agents don't dissolve the problem.
Building ReAct-based AI Agent from Scratch in Ruby programming language
This article explains how to design and implement a simple AI agent architecture in Ruby, focusing on core concepts like perception, reasoning, decision-making, and action rather than just code. Using Ruby's clarity, it demonstrates how modular components come together to form a goal-driven agent that interacts with its environment in a structured, understandable way.
Structured Memory Extraction from Long-Running AI Conversations
Designing and evaluating system-level methods for extracting, updating, and using long-term user memory from conversations—without hallucinated personas, stale assumptions, or excessive compute.
Multi-Objective Reasoning in Agentic LLM Systems
Agentic LLMs struggle to balance competing objectives across long execution horizons. This work studies how models and agent architectures handle persistent, conflicting goals and proposes evaluation methods and design patterns to improve objective-aware reasoning.
Built in the Open
We believe in sharing our tools with the community. These projects power our research and are free for everyone to use.
Regent is a small and elegant Ruby framework for building AI agents that can think, reason, and take actions through tools.
A Hugging Face API client for Ruby
A desktop app for streamlining LLM development lifecycle, from design and testing to documentation and monitoring
Small and elegant TypeScript library for building AI agents that can think, reason, and take actions through tools.
A Small Team with a
Product Mindset
We've spent years building software and now focus on applying AI responsibly — through both paid client work and open initiatives.
Alex Chaplinsky
14+ years experience in building software and AI systems. Specialized in building AI-powered products and open-source tools.
Anton Kostyuchenko
Product visionary focused on transforming businesses with AI solutions, bridging tech innovation with business goals.
