Yash Dagli
Data Scientist @ Microsoft | Building Functional AI Solutions
- Role
- Data Scientist at Microsoft
- Location
- Redmond, WA, US
- LinkedIn followers
- 500 followers
About Yash Dagli
Results-oriented Data Scientist with 4+ years of industry experience and an M.S. in Computer Science. Proven expertise in designing and deploying predictive and generative AI models, automating workflows, and delivering scalable data-driven solutions that drive operational efficiency and business growth. Skilled in ML/DL algorithms, MLOps, AIOps, and GenAI development, with strong proficiency in Python, PySpark, Docker, Agentic Frameworks, and leading cloud-based services.
Experience
Data Scientist
Feb 2022 — Present · Redmond, WA, US
Engineered an end-to-end Azure ML solution to predict uncorrectable DIMM failures for the Azure hardware fleet by developing and fine-tuning stacked XGBoost and LSTM-based model, integrating drift detection, and automating retraining, improving AIR (a key business OKR) by 0.14 and saving over 1,500 developer hours monthly. Spearheaded the design and automation of RAG and Agent based GenAI platforms, building self-service chatbot creation, image retrieval-augmented generation workflows, and CRUD tooling on Vector Database indexes, scaling to 140+ chatbots serving 200+ users with monthly queries and saving 200 developer hours per month. Designed and implemented CAHRR, a correlation engine that analyzes VM interruptions across 4M+ cloud nodes daily, applying statistical modeling (Poisson, Cox regression) to link VM failures with node hardware attributes, enabling intelligent allocation of 1.6M+ VMs to high-rated nodes and reduce Interruption rate by 0.83. Spearheaded the creation and deployment of a custom NLP-based fault classification model (SLAM), incorporating embedding and RNN techniques, achieving ~80% accuracy and reducing operational analysis time by ~30% for hardware triage teams across Azure’s fleet. Highlights:Co-inventor on 4 patents filed at Microsoft.Recipient of Microsoft “Ambassador Award\" in recognition of your exceptional commitment to role-modeling our SCHIE Engineering Principles: People First, Scalable Capabilities, Innovation, Business Results Focus, Ownership Mindset, Trusted Partners.
Education
Northeastern University
Master of Science - MS, Computer Science
2019 — 2021
Ganpat University
Bachelor of Technology, Computer Science & Engineering (Specialization in Big Data Analytics in association with IBM)
2015 — 2019
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.