About Kutup

One mission: purpose-built AI for industrial reality.

Kutup is developed by ASP Dijital as a structured ecosystem of specialized tools. The current production engine, Kutup NQ, is a domain-constrained industrial assistant that combines curated datasets, retrieval pipelines, and guided model synthesis to answer technical questions with visible evidence.

6 focused tools planned Kutup NQ v1.4 (Beta) Curated knowledge + live sources

Core Ideology

One domain, one product Each tool is purpose-scoped so depth is never sacrificed for breadth.

Curated over generated Answers are grounded in maintained datasets first, then expanded through controlled Wikipedia and RSS evidence.

Bilingual by design Turkish and English are both first-class from architecture to interface, including cross-language search expansion.

Inspectable by default NQ exposes source context, tool usage, and evidence panels so engineers can see how an answer was assembled.

Ecosystem Narrative

Kutup is built incrementally: launch one reliable tool, validate it in production, then expand.

Now Live: Kutup NQ

Neural Query is the first production app, focused on industrial automation, IT/OT, ML, protocols, and data-system fundamentals with evidence-aware output.

How NQ Works

NQ routes a query through custom datasets, browser-based ML signals, Qwen3.5-9B synthesis, and selective Wikipedia/RSS enrichment when additional context is needed.

Operational Direction

Modbus LLM, OPC LLM, PLC LLM, Flow Maker, and Kutup Forge extend the platform from technical Q&A into deterministic execution workflows.

Build Principles

The same architecture and UX principles apply across every Kutup application, with NQ as the reference implementation.

Scope boundaries first

NQ rejects irrelevant prompts and only accepts supported English and Turkish industrial-domain requests.

Data curation pipeline

Custom JSON datasets are curated and versioned, then expanded with bilingual retrieval paths to improve technical recall depth.

Production UX for engineers

Session history, thinking states, used-tool chips, and inspectable evidence panels keep the answer flow visible instead of opaque.

Hybrid synthesis layer

Qwen3.5-9B is used as the primary synthesis model, while browser-side ML helps ranking and interaction quality on the frontend.