Technical and operational notes on what we build, test and learn: architectures, automation, local AI, integrations and product decisions.
A model name and a benchmark score are not enough. Before a local model receives production tasks, workload, runtime, constraints and failure modes need to be tested together.
Local LLM · Evaluation · On-premise · Private AI
Running a model inside the company is not automatically safer, cheaper or more useful. It makes sense when control, latency, data ownership and workload justify the infrastructure.
On-premise AI · Cloud · Privacy · Infrastructure
Many automation projects fail because they try to replace the core system. Often a layer that reads, structures, validates and prepares data before it enters the ERP is enough.
Orders · Email · Data entry · Integration
The bottleneck is not always calculating a price. Often it is collecting the information a person needs to make a fast and well-informed decision.
Quoting · Lead qualification · Human review · Workflow
An old or awkward system can still contain essential rules, data and dependencies. Before rewriting everything, identify which part of the work needs a new layer.
ERP · Integration · Custom software · Operations