Saurabh Aggarwal

CEO

Mountain View, California, United States5 yrs 5 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in scaling AI systems under real-world constraints.
Led multi-million dollar projects at Microsoft and Apple.
Strong background in applied machine learning and data science.

Stackforce AI infers this person is a Data Science and AI expert with a focus on cloud solutions and machine learning applications.

Contact

Skills

Core Skills

Artificial Intelligence (ai)Solution ArchitectureData ScienceApplied Machine Learning

Other Skills

Generative AIInference OptimizationMicroservicesCloud ArchitectureAnomaly DetectionMLOpsMarkov Chain ModelingNatural Language ProcessingMachine LearningXGBoostRegression AnalysisContent TaggingExploratory Data AnalysisDeep LearningLSTM

About

Most of my work has happened behind the scenes - across Apple, Microsoft, and now NVIDIA. I’ve spent my career building and scaling AI systems that operate under real-world constraints, where performance, reliability, cost, and scale all matter at the same time. What’s shaped my perspective most isn’t the outcomes, it’s the trade-offs. AI looks very different in production than it does in theory. Systems break. Incentives matter. Constraints shape behavior. If you’re interested in grounded perspectives on AI beyond the hype, feel free to connect or follow.

Experience

Nvidia

Senior Solutions Architect, Cloud AI

Oct 2024 – Present · 1 yr 5 mos · Santa Clara, CA · On-site

[Generative AI (LLM) Inference Optimization, Dissaggregated Serving, Inference Microservices]
I work closely with cloud and platform teams to design architectures that scale across modern GPU systems and real production constraints.
I also share practical lessons from this work through talks and engineering blogs. (Linked below)
This role continues to shape how I think about AI systems not just as models, but as infrastructure that must perform reliably under real-world constraints.

Generative AIInference OptimizationMicroservicesCloud ArchitectureArtificial Intelligence (AI)Solution Architecture

Microsoft

Data Scientist | Azure, Customer Experience Data + AI

Feb 2023 – Oct 2024 · 1 yr 8 mos · Mountain View, California, United States · On-site

[Anomaly Detection, Attribution, Text Tagging, Late Payment Detection, MLOps]
Microsoft is where scale stopped being abstract.
I started as an intern in the Azure DSaaS (Data Science as a Service) team. Worked on:
→ Early warning detection on multivariate time series data
→ Deploying DSaaS offerings on the Azure Marketplace
As a full-time Data Scientist, I worked across Azure, Commerce, Finance, and Microsoft Learn. Some of the work I led were:
→ Daily Cloud Cost Anomaly Detection for all Microsoft customers. Repurposed Google’s WaveNet for large-scale anomaly detection, 91% CSAT and ~$15M in anomalous Azure usage identified daily
→ Cross Cloud Content Attribution on learn.microsoft.com. Used Markov Chain modeling to measure content influence on customer purchases
→ LLM-based multi-label text tagging framework for Microsoft Learn. Leveraged prompt engineering and modern LLMs. Potential to save $50M+ in operational time and costs at scale
→ Late invoice payment detection using XGBoost regression. Recovered $100s of millions in active cash flow. MAE of 0.3 months
Across these systems, I also directed:
→MLOps
→ Security
→ Responsible AI
For 13 production models with a combined influence of over $500M.
I presented 3 papers at Microsoft’s MLADS conferences and published 3 papers in the Microsoft Journal of Applied Research (MSJAR).

Anomaly DetectionMLOpsData ScienceApplied Machine Learning

Apple

2 roles

Machine Learning Engineer

Jul 2019 – Jun 2021 · 1 yr 11 mos

[Content Tagging, De-duplication in Forums, EDA]
At Apple, I worked on applied machine learning for real, user-facing products.
Production systems to be exact.
→ Built models combining probabilistic methods and deep learning to improve content tagging across Apple’s support ecosystem
→ Designed a de-duplication approach using Manhattan LSTMs for the Developer Communities Forum
→ Analyzed Webex usage metadata using exploratory data analysis to uncover behavioral insights
Alongside the technical work, I regularly presented findings to the CIO and senior leadership.
I also mentored interns on presentation skills and professional development.
This was my first deep exposure to building ML systems that needed to be accurate, scalable, and understandable by both technical and non-technical stakeholders.

Content TaggingExploratory Data AnalysisMachine LearningApplied Machine Learning