About
Batuhan Zorbey Zengin leads AI product development and the AI consulting unit at Clerion, an Istanbul-based AWS Advanced Tier Partner. He oversees the development of Torii, Clerion's product that lets companies manage their use of language models and agents inside their own infrastructure, adapted to their own traffic.
His work centres on KV cache, speculative decoding, quantization, RAG architectures and GPU-based model serving. He writes about these topics on LinkedIn and Medium; recent pieces cover agent memory, switching RAG embedding models without reindexing, chunking for enterprise RAG and KV-cache injection between models.
Focus
- KV cache
- Speculative decoding
- Quantization
- RAG architectures
- GPU model serving
- Agent systems
On the Torii blog1
Product & architecture
Why We Built Torii: The Control Problem in Enterprise AI
AI use inside a company grows one team at a time. This is why we built a single control layer for access, cost, data flow and audit, including for the models a company runs itself.
8 min read
Read the postElsewhere
- I Gave My LLM Agent’s Memory a Third FateMedium
- rebasis: Switch RAG Embedding Models Without ReindexingMedium
- Getting Chunking Right: The Most Misunderstood Layer in Enterprise RAGMedium · Clerion
- Digital Telepathy: Redesigning AI Communication with Gemma 3 and KV-Cache InjectionMedium
- API Devrinin Sonu: Yapay Zekalar Neden “İnsan Diliyle” Konuşmayı Bırakmalı?MediumTurkish
Every AI request, priced, capped and logged.
Torii AI Gateway is opening to a first group of teams.