AI & ML interests

E-Commerce

Recent Activity

aiprojeĀ  updated a Space 1 day ago
aiprojecom/README
aiprojeĀ  published a Space 2 days ago
aiprojecom/README
View all activity

Organization Card
AIPROJE

AIPROJE

Turkish datasets, domain-adapted AI models, and a decade of applied engineering — e-commerce is our flagship business, not our whole story.

Website Hugging Face GitHub YouTube X Language


Who we are

AIPROJE was founded as an AI and software projects company, not just an e-commerce infrastructure provider. With a decade of hands‑on experience across AI/ML, software engineering and systems integration, we started by building AI‑powered e‑commerce solutions and have since expanded into developing Turkish language models, datasets and domain‑specific AI for multiple industries. Our platform, aiproje.com, is a comprehensive e‑commerce infrastructure for Turkish B2B, B2C and dropshipping sellers — supporting multi‑language and multi‑currency operations — but our work does not stop there. We also create AI‑powered mobile applications, visual editing tools and tailored AI solutions for sectors ranging from automotive to logistics.

This Hugging Face organization is where we publish our AI research and engineering output:

  • Build Turkish datasets, including Turkish reasoning datasets with step-by-step (think) chains rather than plain answers
  • Fine‑tune open models for concrete, domain‑specific jobs instead of generic chat
  • Optimize inference and adapt models to run on modest hardware — quantization, LoRA, low‑VRAM deployment
  • Work through the decisions behind base model development — architecture, training, and adaptation choices — and share what we learn along the way
  • Feed all of the above back into our commercial products: AI‑assisted product search, translation, image enhancement, pricing, and entirely new AI‑driven applications

Leadership

AIPROJE is led by Doğukan Atakul, who also founded the Bursa AI Community (Bursa Yapay Zeka Topluluğu). His background covers six‑plus years across AI/ML systems, computer vision, model compression and fine‑tuning, low‑resource and edge deployment, domain‑specific model development, and systems integration — alongside the broader software engineering, Linux, and server/network work that the AIPROJE platform itself is built on.

He also publishes Turkish‑language technical videos on his YouTube channel, covering LLMs, fine‑tuning, quantization, MoE, edge AI, and tools like CUDA, PyTorch, TensorRT, vLLM, and Ollama — aimed at explaining AI as engineering rather than hype, and growing the amount of serious Turkish‑language technical content out there.

What's here

Our first release wave centers on Turkish automotive e‑commerce — spare‑parts search, compatibility checks, and technical support — a concrete domain we're using to prove out our approach before generalizing to others.

šŸ¤— Models

Model Base Task What it does
qwen3-1.7b-otomobil-yedekparca-lora Qwen3-1.7B (Unsloth, 4‑bit) Text generation Ā· LoRA A Turkish automotive spare‑parts assistant covering product lookup, pricing, compatibility checks, troubleshooting, function calling, and multi‑turn, step‑by‑step reasoning. Fine‑tuned on tens of thousands of synthetic reasoning QA pairs. By our own benchmark notes it's solid at product search and function calling, weaker at compatibility checks and open‑ended chat — an experimental checkpoint testing how far a small 1.7B model can go in production e‑commerce.
kumru-2b-flirt-lora vngrs-ai/Kumru-2B Text generation Ā· LoRA A small side‑experiment in shifting a Turkish base model's conversational tone toward casual, informal chat. Purely a research/fun project — not intended for production or professional use.

Both models are released under Apache 2.0.

šŸ“š Datasets

Dataset Size License What's in it
turkce-otomobil-yedek-parca-soru-cevap-akil-yurutme ~11,454 rows MIT Our largest release: Turkish reasoning data for real spare‑parts compatibility questions — OEM codes, brands, and model years across dozens of manufacturers — each with a think column showing the reasoning behind the answer.
turkce-sentetik-otomotiv-yedek-parca-akil-yurutme ~990 rows MIT Synthetic Turkish reasoning data on part identification, compatibility, and support‑style conversation, with a deliberate mix of tones and customer personas.
turkce-otomotiv-yedek-parca-youtube-soru-cevap-akil-yurutme ~260 rows MIT Reasoning Q&A distilled from Turkish automotive YouTube content, ranging from maintenance and diagnostics to adjacent engineering topics.
turkce-otomobil-bakim-soru-cevap 1.69 MB MIT A more general Turkish Q&A set on car maintenance and common faults, without the reasoning‑chain annotations of the other three.

All four are Turkish‑language and MIT‑licensed. Combined, the three reasoning sets alone total roughly 12,700 rows — and more reasoning data and checkpoints are on the way.

What we build

Our flagship product, AIPROJE, is a modular e‑commerce platform for retail, wholesale and dropshipping sellers, including exporters. It handles order management, invoicing, stock, and reporting, with multi‑language/multi‑currency support and integrations for local carriers, marketplaces, and payment providers — plus AI features (automatic translation, image enhancement, product copy, market‑based pricing) built directly into the product.

Beyond e‑commerce, we develop AI‑powered mobile applications, image and video editing tools, and domain‑specific AI solutions for industries like automotive, logistics, and retail. These projects often involve creating custom datasets, fine‑tuning models, and deploying optimized inference on edge devices — contributing directly to the Turkish AI ecosystem.

Get in touch

We're glad to hear from researchers, developers, or businesses working on Turkish NLP, e‑commerce AI, or anything adjacent.


Follow this organization for new Turkish models and datasets as they ship.