Vahid Abdollahi
Applied Ai Scientist Ii @Bentley Systems
Signup · Get unlimited contacts
WORK HISTORY
Applied Ai Scientist Ii @Bentley Systems
As a Tech Lead- Architecting an event-driven, microservices-based AI backend on Azure Container Apps, designing a fault-tolerant multi-agent orchestration framework utilizing Temporal for durable execution and Pydantic AI. Standardized tool integration via the Model Context Protocol (MCP) to decouple domain-specialist agents, automating complex IoT engineering workflows and daily planning operations- Owned roadmap definition and designed a modular Azure-based ML orchestration platform for time-series IoT data, aligning UX, frontend, backend, and ML teams to ship features like correlation analysis, confidence-banded regression, and trend detection- Drove product-market fit by translating raw sensor data into predictive insights and intelligent alerting workflows, backed by a scalable backend and a user-friendly interface tailored to industrial monitoring.As an Applied AI/ML Scientist- Created a sensor data linear and non-linear similarity analysis and clustering tool using Principal Component Analysis, Encoder-decoder neural networks, t-SNE, K-means, and hierarchical clustering- Designed and implemented a vibration classification tool by leveraging fast Fourier transform for feature extraction and utilizing LightGBM and Encoder-Decoder LSTM for classification- Investigated chunking strategies for a Retrieval-Augmented Generation (RAG) system tailored to IoT documents, focusing on semantic and agentic chunking approaches
EDUCATION
University of Tehran
BSc, Mechanical Engineering
McGill University
Doctor of Philosophy (PhD), Mechanical Engineering
University of Tehran
MSc, Mechanical Engineering
SKILLS
ABOUT VAHID ABDOLLAHI
Proficient AI/ML Architect with 6+ years of hands-on experience deploying end-to-end machine learning pipelines and 8+ years tackling complex computational engineering challenges. I combine rigor in physics-based simulation with cutting-edge expertise in Machine Learning, Generative AI, and Agentic workflows to solve complex data challenges in engineering.Key Technical Highlights- Agentic AI & Orchestration: Architecting fault-tolerant, multi-agent frameworks using Temporal for durable execution and Pydantic AI. Standardizing external tool and LLM integration via the Model Context Protocol (MCP) to automate high-value engineering workflows- Cloud Architecture & ML Systems: Designing scalable, event-driven microservices deployed on Azure Container Apps to process real-time telemetry and schedule automated planning tasks- Deep Learning & RL: Advanced work in Reinforcement Learning, Behavior Cloning, LSTMs, Transformers, and Computer Vision for time-series forecasting and decision-making- Data & Simulation: Expert in synthetic data generation, reduced-order modeling, and high-performance computing (MPI).Leadership & Collaboration- Proven technical leader for cross-functional teams, translating complex business requirements into scalable architectures for large-scale monitoring platforms- Experienced in mentoring junior engineers, evaluating cutting-edge orchestration frameworks, and driving AI/ML roadmaps from concept to production in Agile environments.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.