Spaces:
Running
Running
| title: README | |
| emoji: 🐠 | |
| colorFrom: purple | |
| colorTo: indigo | |
| sdk: static | |
| pinned: false | |
| # TankDev | |
| **TankDev is a software and AI engineering company based in Türkiye, building custom software systems, applied AI solutions, process automation, and system integrations.** | |
| We design software around real operational workflows — from data models and business rules to APIs, integrations, automation, and production deployment. | |
| ## Engineering Areas | |
| - **Custom Software Development** — Business-specific software systems built around operational requirements and workflows. | |
| - **Applied Artificial Intelligence** — AI models and AI-assisted systems for classification, analysis, computer vision, and decision support. | |
| - **Process Automation** — Traceable workflows that automate repetitive operational tasks and system actions. | |
| - **System Integration** — Reliable data exchange between ERP, CRM, accounting, web applications, APIs, and other business systems. | |
| - **Web & Backend Engineering** — Production-oriented applications, APIs, databases, authentication, background jobs, and integrations. | |
| - **Software Modernization** — Migration of legacy applications and fragmented workflows into maintainable software architectures. | |
| ## Research & Engineering Projects | |
| ### DeepCancer | |
| DeepCancer is TankDev's research project exploring artificial intelligence for histopathology image analysis. | |
| The current research prototype investigates breast cancer histopathology using hybrid deep-learning architectures and explainability techniques such as Grad-CAM. | |
| DeepCancer is intended for research and education and is **not a clinical diagnostic tool**. | |
| [](https://deepcancer.org) | |
| ### Autonomous AI Video Studio | |
| An automated production system that connects research, AI-assisted content generation, media production, validation, and publishing into a controlled workflow. | |
| ### AI Automation Reliability | |
| TankDev also researches reliability in multi-step AI and LLM workflows, including failure propagation, retries, validation, observability, and measurable end-to-end reliability. | |
| ## Engineering Principles | |
| We focus on systems that remain understandable and recoverable when something goes wrong. | |
| Our engineering work emphasizes: | |
| - explicit data ownership | |
| - API contracts | |
| - idempotency and safe retries | |
| - traceable workflows | |
| - validation and acceptance criteria | |
| - observability | |
| - failure recovery | |
| - measurable AI performance | |
| - maintainable system architecture | |
| ## Technologies | |
| Our engineering work commonly involves: | |
| `Python` · `FastAPI` · `PostgreSQL` · `Next.js` · `TypeScript` · `Docker` · `Linux` · `Machine Learning` · `Computer Vision` · `LLM APIs` | |
| Technology selection depends on the requirements and constraints of each system. | |
| ## TankDev on the Web | |
| **Website:** https://tankdev.tech/en | |
| **Engineering Blog:** https://tankdev.tech/en/blog | |
| **Systems & Research:** https://tankdev.tech/en/systems | |
| **About TankDev:** https://tankdev.tech/en/about | |
| **Contact:** https://tankdev.tech/en/contact | |
| --- | |
| **TankDev — Software & AI Engineering** | |
| Custom Software · Applied AI · Process Automation · System Integration | |