Kartik Mathur — AI Researcher

My views are my own and do not reflect my employer. I primarily work in Python, and have experience across the modern ML stack, including data processing, model development, and system integration. I am an Applied Machine Learning Scientist at Microsoft, where I design, build, and deploy production-grade AI systems that power user-facing experiences across Office 365 at scale. My work focuses on Large Language Models (LLMs), agentic systems, and applied ML, with an emphasis on reliability, evaluation, and measurable real-world impact. I specialize in taking models from early experimentation to full production ownership. This includes problem formulation, offline evaluation, online A/B experimentation, deployment, monitoring, and iteration. I have hands-on experience building and operating LLM-based systems, including model serving and inference pipelines, latency and cost optimization, and production observability. I care deeply about evaluation rigor, including LLM benchmarking, hallucination mitigation, and aligning offline metrics with online outcomes. My background spans applied ML, experimentation platforms, and data-informed decision-making in distributed production environments. I have worked extensively with large-scale datasets, feature pipelines, and cloud-native ML infrastructure, partnering closely with engineering, product, and design teams to ship AI-driven features under real-world constraints. My interests include RAG systems, tool and function calling, multi-agent orchestration, LLM safety and guardrails, and scalable evaluation frameworks. Beyond product work, I actively contribute to the broader research and engineering ecosystem. I am an AAAI author and Program Committee member, a patent inventor, and a hackathon winner, reflecting a sustained focus on both innovation and execution. I enjoy bridging applied research with production systems and translating ambiguous problems into deployable solutions. I hold a Master’s degree in Computer Science from the University of Southern California and bring a systems-oriented, pragmatic approach to applied AI, grounded in theory but driven by deployment realities. I value technical ownership, participate in design and architecture discussions, and enjoy mentoring and collaborating across disciplines. I am always open to thoughtful conversations around LLMs, agentic AI, applied ML research, experimentation, and production-scale AI systems, as well as opportunities to collaborate on impactful problems.

Stackforce AI infers this person is a SaaS-focused Applied Machine Learning Scientist with expertise in AI systems and large-scale deployments.

Experience: 6 yrs 2 mos