Tariful Islam Tarif

AI/ML Engineer driving the design, deployment, and operation of production-grade machine learning, LLM, and data-driven systems. Expertly crafting sca...

About

AI/ML Engineer driving the design, deployment, and operation of production-grade machine learning, LLM, and data-driven systems. Expertly crafting scalable, low-latency inference pipelines, multi-agent LLM architectures, and end-to-end MLOps workflows with Python, FastAPI, Docker, and Kubernetes.

Skills

Programming Languages

Python SQL C++ Java C

Machine Learning & Deep Learning

PyTorch TensorFlow Scikit-learn ONNX Runtime Model Training & Evaluation Model Versioning

LLMs & Generative AI

LangChain LangGraph Retrieval-Augmented Generation (RAG) Prompt Engineering Multi-Agent Systems LLM-as-Judge Evaluation OpenAI API Hugging Face

Data Science

Data Mining Statistical Analysis Exploratory Data Analysis NL-to-SQL Feature Engineering

Backend & APIs

FastAPI REST API Design Asynchronous Programming Microservices Architecture

Databases

PostgreSQL MongoDB SQLite Vector Databases (FAISS, Pinecone, Chroma)

MLOps & Observability

Prometheus OpenTelemetry Structured Logging Drift Monitoring Model Versioning

DevOps & Tools

Docker Kubernetes Git GitHub Canary Rollouts Autoscaling

Projects

Intelligence Reliability Platform — LLM Observability & Evaluation

FastAPI React OpenAI API Ollama SQLAlchemy Docker

Built a production-grade LLM observability platform featuring prompt versioning, risk controls, and multi-model fallback routing (OpenAI → Ollama → Mock) for high availability. Designed automated LLM evaluation pipelines using ensemble LLM-as-judge scoring across quality, safety, and hallucination-detection dimensions. Implemented real-time trace logging, cost tracking, latency metrics, and drift monitoring to support continuous model governance. Developed role-based access control (RBAC), audit logging, and secure inference workflows for enterprise-grade compliance.

Real-Time ML Inference System

FastAPI ONNX Runtime Redis Prometheus Docker Kubernetes

Engineered a low-latency ML inference API supporting multi-model and multi-version serving. Achieved sub-100ms inference latency using ONNX Runtime and Redis-backed intelligent caching, cutting redundant compute by 80%+. Implemented Prometheus metrics, health checks, and OpenTelemetry-based structured logging for full-stack observability. Deployed via Docker Compose and Kubernetes with autoscaling and canary rollout support for zero-downtime releases.

Multi-Agent Medical Data Assistant

Python Google Gemini GPT-4o-mini SQLite NL-to-SQL

Built a multi-agent LLM system enabling natural-language querying of medical datasets (heart disease, cancer, diabetes). Implemented NL-to-SQL pipelines with automated routing between database query tools and medical web-search agents. Integrated LLM-based summarization to generate structured, evidence-based analytical insights for healthcare data.

Advanced Image Change Detection System

Python OpenCV ORB SSIM

Designed a computer vision pipeline for robust before/after image comparison under camera misalignment. Implemented ORB feature matching, homography alignment, and SSIM-based structural change detection. Generated annotated outputs, heatmaps, binary masks, and structured JSON summaries to support auditability.

LangGraph-Style Multi-Agent Debate Workflow

Python LangGraph Graphviz

Built a deterministic multi-agent orchestration system with strict turn control and debate sequencing logic. Implemented memory nodes, semantic duplicate detection, and replayable JSON-based event logging. Visualized multi-agent workflows as DAGs to support debugging and evaluation.

Experience

Forward Deployed Engineer

INFINOZ

01/03/2026 - Present

Partnered directly with product and client teams as a Forward Deployment Engineer to design, build, and deploy applied AI solutions from research through to production. Architected and developed InfiChat, an AI-driven conversational assistant, owning model integration, system architecture, and end-to-end deployment. Led R&D on a real-time Voice Agent, integrating speech recognition (ASR), natural language understanding (NLU), and speech synthesis (TTS) into a unified conversational pipeline. Fine-tuned and evaluated Bangla Text-to-Speech (TTS) models, covering dataset preparation, model fine-tuning, and voice-quality evaluation to improve native Bengali speech synthesis. Collaborated cross-functionally with engineering and product stakeholders to convert experimental AI research into reliable, scalable, production-ready voice and chat-based AI components.

Education

B.Sc. in Data Science & Engineering

University of Frontier Technology Bangladesh (UFTB)

- Present

Relevant Coursework: Machine Learning, Deep Learning, Natural Language Processing, Data Mining, Statistics, Python Programming

Certifications

AI Engineer Bootcamp

Ostad

Contact

Email: tarifulislamtarif961@gmail.com