Tariful Islam Tarif

AI and Machine Learning Engineer with hands-on experience designing, deploying, and operating production-grade ML and LLM systems. Strong focus on reliability, scalability, and real-time inference, with experience building observable, testable, and reproducible AI services using FastAPI, Docker, and Kubernetes-ready architectures. Demonstrated ability to translate product and reliability requirements into measurable metrics such as latency, cost, availability, and model quality.

Experienced in LLM evaluation, multi-agent orchestration, monitoring, and model versioning, with a solid academic foundation in Data Science & Engineering (CGPA: 3.54/4.00).

Forward Deployment Engineer - INFINOZ

(2026-03 - 2026-07)

Worked as a Forward Deployment Engineer at INFINOZ, partnering directly with product and client teams to design, build, and deploy applied AI solutions from research to production.

• Researched, designed, and developed InfiChat, an AI-driven conversational assistant, covering architecture design, model integration, and end-to-end deployment.
• Led research and development of a real-time Voice Agent, integrating speech recognition, natural language understanding, and speech synthesis into a unified conversational pipeline.
• Conducted focused research and development on Bangla Text-to-Speech (TTS) models, working on dataset preparation, model fine-tuning, and voice quality evaluation to improve native Bengali speech synthesis.
• Collaborated cross-functionally to translate experimental AI research into deployable, production-ready components, ensuring reliability and scalability of voice and chat-based AI systems.

B.Sc. - Data Science & Engineering - University of Frontier Technology Bangladesh (UFTB) (2024)