AI Design
A solution shaped by the business task. Cost per request is calculated before development starts.
Read more about AI DesignSystems in production
AI infrastructure, agents & IT consulting
We design, build and run our own systems. Every one of them is monitored, has a rollback plan and a known cost per request.
- 5
- systems live
- 24/7
- monitored
- 1
- rollback plan per change
What we do
AI Development
Gateways, agents, RAG, integrations with CRM and ad platforms. Monitoring and a rollback plan included.
Read more about AI DevelopmentAI, Software & IT Consulting
An audit of what already runs: where AI pays off, where it is redundant, what it costs to operate. The same rigour for software and IT decisions.
Read more about AI, Software & IT ConsultingRunning in production
Proxy Broker
liveAI gateway: routing by model alias, failover on 429 and 5xx, circuit breaker, metrics in Grafana.
- Python
- PostgreSQL
- Redis
RAG Platform
on-premiseVector search on your own hardware: Qdrant and local LLMs. Data never leaves the premises.
- Qdrant
- Ollama
- MCP
Ads Automation
marketingCampaign audit and optimisation through ad platform APIs. Conversion tracking is verified in GA4 first.
- Ads APIs
- GA4
- Python
ClaudeClaw
agentsA bridge between Telegram and a developer agent: multi-agent orchestration and a three-layer memory.
- Agents
- FTS5
- Telegram
BookWriter
in progressA book assembled from a corpus of an author's posts. A quality gate removes AI clichés; the output is EPUB plus narration.
- Multi-agent
- QA gate
- TTS
Quality
Code and architecture go through review. Tests and monitoring from day one, before the first incident rather than after it.
Availability
The loss of one provider or node does not take the system down: reserve capacity and failover are part of the design.
Reliability
Every change has a rollback plan. Backups are made and verified by restoring them.
Tell us about the task
We will look at where AI pays off and where it would cost more than it returns.