Educational AI engineering videos from our team: agent harnesses, orchestration, context engineering, eval-driven development and the architecture decisions behind production LLM systems.
Subscribe on YouTubeMirko Solutions Media
Short, technical episodes from our engineers on agent harnesses, orchestration, context engineering and evaluation. Press play to watch here, or open the channel on YouTube.
September 2026
How to build reliable evaluation pipelines for production LLM applications, including automated testing, LLM-as-a-Judge frameworks, benchmarks and regression detection.
September 2026
A technical comparison of fine-tuning, Retrieval-Augmented Generation (RAG) and AI scaffolding. Explains how to choose the right architecture based on total cost of ownership, latency, adaptability and performance.
August 2026
How to transform a stateless language model into a persistent AI worker using execution tools, memory, task management, verification and secure operational infrastructure.
August 2026
Four categories of AI agent execution environments, from interactive coding assistants to fully autonomous server-based agents. Compares architecture, execution models, security and operational constraints.
August 2026
How managers can build persistent personal AI workspaces with Claude Projects and ChatGPT without custom software: instructions, knowledge management, memory and assistant configuration.
July 2026
Practical multi-agent orchestration architectures: LangGraph-based workflows compared with asynchronous Kanban-style coordination. State management, parallel execution, failure recovery and human verification.
July 2026
A technical overview of context engineering for LLM applications: context-window management, token budgeting, memory strategies, context pruning and attention optimization.
July 2026
The shift from manual programming to specification-driven AI development, where engineers define precise technical requirements and AI agents execute implementation under automated test constraints.
July 2026
Security architecture for autonomous AI agents: sandboxed environments, granular permission controls, human approval checkpoints and safeguards against unintended actions.
Have a project in mind or looking for the right technology partner? Tell us what you're working on, and our team will get back to you to explore how we can help bring it to life.