Search and ranking
Relevance models for product search, autocomplete and recommendations; test sets and graded judgements to measure whether ranking improved.
AI engineering · Gothenburg, Sweden
AI systems that hold up in production.
We build AI systems: search and ranking, agents and tool integrations, fine-tuned and self-hosted models. And we audit software for stability and security, fix what we find, and stand behind the result with a written warranty.
- Fixed scope, written quote
- Senior-engineer hours, stated up front
- Warranty in fix hours
AI ENGINEERING
AI that works outside the demo.
Agents and tool servers
Chat assistants grounded in your own data and tools; MCP servers so ChatGPT and Claude can use your systems safely, with rate limits and scoped access.
Evaluation and LLM judges
Benchmarks built from your real tasks; small judge models that grade outputs consistently and cheaply.
Model development
Fine-tuning small open models with LoRA for a narrow task, trained on our own GPUs or yours; we ship the evaluation alongside the model.
Self-hosted deployment
Open models served on hardware you control, for data that must not leave your network.
SOFTWARE AUDITS
Software audits with a warranty.
Stability, security and code-quality review of your application, with a prioritised fix plan, a re-check of your fixes and a written warranty.
How an audit works
-
Audit
Fixed fee. Clear findings.
-
Fix
Your team or ours.
-
Re-review
One round included.
-
Sign-off
The warranty starts.
-
Retainer, optional
Keep new releases in check.
Human accountability
Honest about AI
Our work is heavily AI-assisted, and we say so. Every quote states the senior engineer hours we commit to your code, a senior engineer reviews every finding, and our warranty is measured in the same hours.
Our AI-assisted processStart a conversation