I build applied AI systems for operational workflows, with a focus on multimodal agents, tool-using LLM applications, and Python backend systems.

My strongest recent project is Maintenance-Eye: a real-time maintenance copilot that combines live audio, camera input, tool use, and human-in-the-loop confirmation for field technicians. I built it with Google ADK, Gemini Live API, FastAPI, Firestore, Cloud Run, Docker, and Terraform.

I hold a Master of Data Science from UBC and currently work at BC Rapid Transit Company, where I build data pipelines and prototype maintenance-focused AI workflows. My portfolio is centered on shipped, operationally grounded systems rather than AI demos without deployment or workflow context.

What I Build

  • Applied AI Systems — multimodal agents, tool-using LLM applications, structured retrieval, and human-in-the-loop workflows
  • Python Backends & APIs — FastAPI, Django, data pipelines, SQL, and production-minded service design
  • Cloud Delivery — GCP, AWS deployment workflows, Docker, Terraform, CI/CD, and production-minded application delivery
  • Operational Data Workflows — maintenance analytics, ETL pipelines, transformation layers, and data modeling for decision support

Experience

BC Rapid Transit Company — Data Analyst

Dec 2024 - Present | Burnaby, BC

  • Build Python- and SQL-based data pipelines for maintenance and asset data, supporting reliability analysis and operational reporting
  • Develop analytics workflows for maintenance planning, KPI tracking, and engineering decision support, including modeling and transformation work in Microsoft Fabric
  • Prototype AI-assisted maintenance workflows that combine camera input, natural-language interaction, and structured asset context for technician diagnostics

Statistics Canada — Data Scientist (Capstone)

Apr 2023 - Jul 2023 | Ottawa, ON

  • Implemented clustering algorithms (K-means, hierarchical) to segment the Proximity Measure Database for urban planning applications
  • Conducted data profiling, exploration, and outlier detection in R, then translated the findings into cluster profiles for policy use

Softology IT — Co-Founder

Sep 2021 - Aug 2022 | Dhaka, Bangladesh

  • Co-founded a software development firm delivering Android apps, e-commerce solutions, and UI/UX design for local clients
  • Managed end-to-end project delivery across multiple concurrent client engagements under tight client and timeline constraints

Education

University of British Columbia — Master of Data Science (Sep 2022 - Jun 2023)

American International University-Bangladesh — BSc in Computer Science & Engineering, CGPA: 3.91/4.0 (Jan 2018 - Sep 2021)

Certifications

Skills

Applied AI & LLM Systems: AI Agents, LLM Integration, Tool Use, RAG, Prompt Engineering, Evaluation, Multimodal AI, NLP, Computer Vision

Backend & Data: Python, FastAPI, SQL, PostgreSQL, Django, Pandas, NumPy, ETL Pipelines, Data Modeling, Microsoft Fabric

Cloud & Delivery: GCP (Cloud Run, Firestore, Gemini), AWS deployment workflows, Docker, Terraform, GitHub Actions, CI/CD, REST APIs

Analytics & ML: Scikit-learn, TensorFlow, R, Power BI, Tableau, Classification, Clustering