About TutanicProduction AI software brought to life by elite Egyptian engineering. AI so advanced they'll blame aliens again.
About the RoleYou'll help build the AI backbone of an AI-powered CRM and case management platform used in real operational workflows. This role is focused on applied ML, not research turning messy real-world data (documents, forms, conversations) into reliable, production-grade AI systems.
You'll work closely with the CTO and backend team to design, implement, and ship AI features that are accurate, debuggable, and scalable in production.
What You'll Work On- Document intelligence pipelines (OCR, extraction, classification, validation)
- Semantic search and retrieval over structured and unstructured data
- Model inference services integrated into backend APIs
- Monitoring, evaluation, and iteration on live AI systems
Responsibilities- Build, train, and deploy ML models for document understanding and automation
- Design data pipelines for training, evaluation, and inference
- Integrate AI services into backend systems (APIs, queues, async workflows)
- Improve model reliability, latency, and accuracy in production
- Collaborate with backend, product, and AI engineers to deliver end-to-end features
Requirements- Strong Python experience with applied ML libraries (PyTorch, TensorFlow, Hugging Face)
- Experience building production ML systems, not just notebooks
- Solid understanding of data preprocessing, evaluation, and failure modes
- Hands-on experience deploying or integrating models with backend services
- Comfortable working with imperfect data and evolving requirements
Nice to Have- Experience with LLMs, embeddings, or retrieval-augmented generation (RAG)
- Exposure to AWS AI infrastructure (SageMaker, Bedrock, Triton, Lambda)
- Familiarity with async systems, queues, or event-driven architectures
- Some backend experience (Node.js / APIs) is a plus, not a requirement
