Senior AI Engineer · Ankara, Türkiye

Website · LinkedIn · ORCID · Contact

Five years building production AI systems — LLMs, RAG, agentic platforms, multi-tenant NER, and the Kubernetes infrastructure underneath. Background spans fintech and healthcare. Published researcher in deep learning (Kolmogorov–Arnold networks, hybrid CNN architectures). Currently pursuing M.Sc. in Computer Engineering at Ankara University.

• Production LLM and agentic-AI systems at Etiya
• M.Sc. in Computer Engineering at Ankara University
• Writing about production AI on berkkirik.github.io/blog

• Senior AI Engineer — Etiya (Sep 2025 – Present) · Türkiye
• Data Scientist Team Lead / Backend Engineer — Im-Par (Sep 2023 – Sep 2025) · Frankfurt am Main, Germany
• Data Scientist — Tiga Information Technologies Inc. (Aug 2021 – Sep 2023) · Ankara, Türkiye

• Enterprise RAG System — Hybrid search (vector + BM25) over 500K+ docs with HyDE rewriting and cross-encoder re-ranking. 92% recall@10, sub-3s responses.
• Agentic AI Platform — Multi-agent orchestration (LangGraph + CrewAI) across 25+ tools with episodic + knowledge-graph memory. 99.2% task completion.
• LLM Guardrail System — Multi-layer (Guardrails AI + NeMo + custom) with 97% PII detection and SHAP explainability. 89.2% blocked, 5K+ concurrent.
• Multi-Tenant NER Platform — Tenant-isolated NER with SLA-based model selection. Sub-200ms inference.
• Code-Generation Fine-tuning — LoRA on Hugging Face code models with A/B adapter swapping. +40% accuracy, −90% training cost.

Languages: Python · Go · C++ · PHP · SQL ML / Deep Learning: PyTorch · TensorFlow · Keras · JAX/Flax · Hugging Face · LoRA · XGBoost LLM / Agentic Systems: LangGraph · CrewAI · LangSmith · RAG (FAISS, BM25, HyDE) · RAGAS · Guardrails AI · NeMo Guardrails Backend & APIs: FastAPI · Django · Flask · Node.js · gRPC · REST · KrakenD Data & Graphs: PostgreSQL · Oracle · Redis · Kafka · Neo4j · GraphQL · Pandas · Spark Infra & DevOps: Kubernetes · Docker · Docker Swarm · GitOps · AWS · Prometheus/Grafana · Keycloak Biomedical / Domain: Signal processing (FFT, PCA, LDA) · Sensor fusion · Embedded (ESP-32, Raspberry Pi) · Healthcare AI

• B. Kırık, G. Uğurlu, A. Aydın, F. Ekinci, K. Açıcı, E. Kumru, A. F. Korkmaz, M. Sevindik, M. S. Güzel, and I. Akata. "Interpretable mycology: leveraging Kolmogorov–Arnold networks for high-accuracy Lactarius species classification and comparative benchmarking." The Journal of Supercomputing, Springer, 2026. First author.
• Y. Atilkan, B. Kırık, E. T. Açıkbaş, F. Ekinci, K. Açıcı, T. Aşuroğlu, R. Benzer, M. S. Güzel, and S. Benzer. "Enhancing crayfish sex identification with Kolmogorov–Arnold networks and stacked autoencoders." Scientific Reports, Nature, 2025.
• Y. Atilkan, B. Kırık, K. Açıcı, R. Benzer, F. Ekinci, and M. S. Güzel. "Advancing Crayfish Disease Detection: A Comparative Study of Deep Learning and Canonical Machine Learning Techniques." Peer-reviewed journal, 2024.

Full publication record on ORCID.

• M.Sc. Computer Engineering — Ankara University (Feb 2026 – Present)
• B.Sc. Biomedical Engineering — Ankara University (Aug 2018 – May 2024)
• Associate Degree, Computer Programming — Istanbul University (Aug 2022 – Jul 2024)

Ankara, Türkiye · berk.kirik@outlook.com Full CV, case studies, and notes at berkkirik.github.io