BLOG · FLOW & AGENTS
Experiments thatare worth telling.
We test, build and document what we learn working with artificial intelligence and real systems.
From a simple task to a job well done
How a chain of AI agents turns building an Agile retrospective into a professional, visual and actionable experience without bundling all the context into a single assistant.
“The key is not doing everything, but doing it better, with purpose and the right tools.”
From One Language to Three:How an Agentic Team Multiplied Our Blog with Hermes, GitOps and a Local Model
How an agentic team (Hermes + GitOps + OpenCode + a local Qwen 3.8 27B) took our Astro blog from one language to three in a week, with Kanban tasks, human review and technical detail you can replicate.
“The key is not doing everything, but doing it better, with purpose and the right tools.”
Agile Is Not Running Faster:It Is Learning First
Agile + AI: the fundamentals you should master before integrating artificial intelligence into your team, with practical examples from Scrum, Kanban and tooling.
“The key is not doing everything, but doing it better, with purpose and the right tools.”
ROADMAPS
Our next practical cases
We build, test and document. These are the next experiments we will share step by step.
- Published
- Upcoming
- On the roadmap
BUILD · AI & INFRA
ROADMAP 1Practical cases about AI, infrastructure and everything we build to bring models into real environments.
- Kubernetes
- Homelab
- Self-hosted
- vLLM
- Models
- Security
- MCP
- CI/CD
- Observability
Getting started with Kubernetes
Everything we learned standing up our first cluster and moving the infrastructure to the homelab.
Setting up our core
The base services that hold the house together: networking, storage and the essential pieces of our homelab.
Deploying our first model with vLLM
Our first real deployment of a local model with vLLM: hardware, configuration and early results.
Securing the model
Authentication, networking and best practices to expose a local model without opening the front door.
Connecting the essential MCPs
The first MCPs we connected to our agents and what changed in the workflow.
Automating with CI/CD
How we stopped repeating tasks: automated pipelines and deployments for our whole stack.
Orchestrating everything with Hermes
The orchestrator that coordinates agents, tasks and deployments so everything runs on its own.
FLOW · TEAMS & AGENTS
ROADMAP 2Use cases about agility, teams and agents to improve how we work, both in ongoing and project work.
- Agile
- Teams
- Productivity
- Ongoing
- Projects
- Dynamics
- Agents
Agents that support the team
How we use agents to automate tasks, answer questions and free up the team's time.
Scrum in action: from plan to impact
Real cases of how we apply Scrum to deliver value iteratively.
Managing ongoing work
Prioritization, focus and steady flow to stay on course and improve every day.
Dynamics that drive the team
Retrospectives, feedback and actions that strengthen collaboration and trust.
Align, communicate and act
Tools and practices to communicate better, align expectations and move the team into action.