Sheyda Kiani-Mehr
Machine Learning Engineer | Data Scientist | Ph.D. Computer Science | GenAI | LLM | AI/ML | Agentic AI | Video QoE | CDN
- Role
- Machine Learning Engineer Data Scientist at Ericsson
- Location
- San Francisco, CA, US
- LinkedIn followers
- 500 followers
About Sheyda Kiani-Mehr
I’m a Machine Learning Engineer and Data Scientist with a PhD in Computer Science from the University of Missouri–Kansas City. My expertise spans large language models (LLMs), natural language processing (NLP), and predictive modeling for video streaming Quality of Experience (QoE).My PhD research focused on optimizing video streaming through cache prefetching and QoE prediction—bridging theory and real-world systems to enhance user experience and streaming efficiency. Since then, I’ve applied AI and machine learning across NLP pipelines, Generative AI (including RAG systems), and time series forecasting.I’m passionate about solving complex problems using deep learning, scalable ML pipelines, and production-ready solutions. I thrive at the intersection of research and impact—transforming ideas into deployed, data-driven systems.
Experience
Machine Learning Engineer Data Scientist
Jun 2021 — Present · San Francisco, CA, US
NLP & Recommendation Systems – Contributed to NLP model development on customer-support logs for recommending solution packs. Involved in data labeling, Transformer-based embeddings, NER, keyword and topic modeling, and applying active learning to improve model quality in collaboration with support teams.• Video Streaming QoE & Time-Series Forecasting – Developed classification models to predict video Quality of Experience (QoE) and applied LSTM / Prophet for network throughput and latency forecasting. Collaborated on Redis-based data processing, feature engineering, class imbalance handling, and anomaly detection to support network analytics.• Generative AI Chatbot – Worked on a Generative AI–based remedy finder using large language models, Retrieval-Augmented Generation (RAG), LangChain, and vector databases to support real-time recommendations.• Production Deployment – Supported model deployment by collaborating with DevOps to containerize components using Docker and integrate into Kubernetes-based production systems.• Cross-Functional Collaboration – Partnered with product management, customer support, and engineering teams to refine model outputs, gather feedback, and align technical priorities with business goals.• Knowledge Sharing – Assisted junior team members with data preparation and model evaluation techniques; contributed to internal documentation and shared best practices across the ML team.
Education
University of Missouri-Kansas City
PhD
2016 — 2021
University of Missouri-Kansas City
Master of Science - MS
2014 — 2016
Skills
- Dash
- Geni
- Shell Script
- Ip Multimedia Subsystem
- C
- Programming
- Cloud Benchmarking and Analysis
- Visual Studio
- Cloud
- Web Development
- Ampl and Cplex
- C++
- Cdn
- Pyspark
- Microsoft Sql Server
- J2me
- Network Routing and Switching
- Css
- Video Streaming
- Borland C++
- X86 Assembly
- Iot
- Cloud Architecture
- Cloud Lab
- C#
- .net
- Oop
- Network Protocols
- Ngn
- Amazon Web Services (Aws)
- Windows
- Traffic Engineering
- Windows Server
- Visual C#
- Html
- Network+
- Google App Engine
- Packet Tracer
- Python
- Asp
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.