Jayant Agrawal

AI Researcher

Vancouver, British Columbia, Canada7 yrs 11 mos experience
Most Likely To SwitchAI Enabled

Key Highlights

  • Expert in developing AI-driven content recommendation systems.
  • Proven track record in leading cross-functional teams.
  • Skilled in building robust evaluation frameworks for AI models.
Stackforce AI infers this person is a highly skilled AI and Machine Learning professional with a focus on recommendation systems.

Contact

Skills

Core Skills

Machine LearningDeep LearningRecommendation SystemsNatural Language Processing

Other Skills

CC++ClusteringComputer VisionConvolutional Neural NetworksEvaluation FrameworksMultilingual ModelsObject DetectionPrompt EngineeringPythonTopic Extraction

About

Developing LLM Agent Architectures for Copilot Discover - a hyper personalized AI Content Feed - that goes from interest/topic level relevance to goal/intent level relevance Previous: User Profiling / Understanding, Recommendation Systems for Cold Users, SLMs for Fine-Grained Topic Extraction

Experience

7 yrs 11 mos
Total Experience
3 yrs 11 mos
Average Tenure
4 yrs 9 mos
Current Experience

Microsoft

3 roles

Senior Applied Scientist - Copilot Discover, Microsoft AI

Promoted

Sep 2025Present · 9 mos · Vancouver, British Columbia, Canada

  • Designed and implemented Agent workflows for Copilot Discover - a hyper-personalized and pro-active AI Content Feed, built robust multi-agent RAG systems leveraging prompt engineering and custom-built automated prompt-tuning frameworks
  • Defined the evaluation strategy for the product across multiple quality dimensions; built an end-to-end evaluation framework encompassing dataset strategy and sampling, LLM-as-a-judge evaluators with strong human alignment, and proactive red-teaming and safety setups to ensure robustness and responsible deployment at scale.
Machine LearningDeep LearningPrompt EngineeringEvaluation Frameworks

Senior Applied Science Manager - MSN Vertical Content Recommendation

Jul 2023Sep 2025 · 2 yrs 2 mos · Vancouver, British Columbia, Canada

  • Led a cross-functional team of Applied Scientists and Software Engineers, owning hiring and career development. Accountable for delivering business-critical outcomes, including incremental DAU through personalized news experiences.
  • News Headlines: Developed a near–real-time clustering pipeline processing ~1M articles per day, incorporating embedding-based clustering and GraphRAG-based summarization to incorporate structured data such as finance/sports, and LLM-as-a-judge evaluation pipelines.
  • Developed and fine-tuned a suite of high-quality news text understanding small language models, including entity and topic extraction, sentiment analysis. Built and scaled custom multilingual SLMs and Entity-Linking pipelines to support high-QPS production workloads.
Recommendation SystemsClusteringNatural Language ProcessingMultilingual Models

Senior Applied Scientist - MSN Finance

Sep 2021Jul 2023 · 1 yr 10 mos · Vancouver, British Columbia, Canada

  • Document Text Understanding - Zero-Shot Fine-Grained Topic Extraction with SLMs
Natural Language ProcessingTopic Extraction

Jumio corporation

2 roles

Machine Learning Engineer

Sep 2018Aug 2021 · 2 yrs 11 mos

  • Structured Data Extraction from identity document images; Real-time Video Selfie to ID Verification
Machine LearningComputer Vision

Machine Learning Intern

May 2018Aug 2018 · 3 mos

  • Digital Manipulation Detection on low quality images with CNNs
Convolutional Neural Networks

Ibm research

Research Intern

May 2017Jul 2017 · 2 mos · Bengaluru Area, India

  • Real-Time Object Detection with single-shot NN on low-compute devices
Object Detection

Education

Indian Institute of Technology, Kanpur

Bachelor of Technology - B.Tech — Computer Science

Jan 2014Jan 2018

Little Angels High School

High School

Stackforce found 100+ more professionals with Machine Learning & Deep Learning

Explore similar profiles based on matching skills and experience