AI writes the questions
From your manuals, policies or course notes, the model drafts a question bank graded by Bloom level. Trainers can review and edit it.

RASEEKH · AI KNOWLEDGE RETENTION
Upload the material and RASEEKH’s AI builds a Bloom-graded question bank from it. Every working day each learner gets one or two short questions, answers in their own words, gets feedback from the AI within seconds, and is asked again just before they would forget.
WHAT RASEEKH DOES
Open-source core: FastAPI · PostgreSQL · OpenSearch · local LLM. Delivered, handed over and supported by Alin Li, and able to carry your own product name.

AI INSIDE RASEEKH
RASEEKH is built around AI. Every step a learner sees is prepared, graded or explained by a model running on your own server.
From your manuals, policies or course notes, the model drafts a question bank graded by Bloom level. Trainers can review and edit it.
Answers in the learner’s own words are graded in seconds against your material, with feedback on what was missed.
The Knowledge Score comes from graded answers over weeks, not from clicks or completed videos.
A local open model does the work on one host. Nothing is sent to a cloud AI service.
PART OF ONE FAMILY
Every product is managed from the Alin Li Platform Console and shares sign-in, audit and health.

SOC and SIEM
A sovereign SIEM and XDR security operations centre on an open core.
Wazuh · OpenSearch
Endpoint log retention
An agent per device, a collector and an index. Every event kept, and restored on request.
Vector · OpenSearch
SOAR and automation
Playbooks that act on what the SIEM finds.
ShuffleSTAY CURIOUS
A short demo with an engineer, then a time-bounded proof of concept on your own data.