AI, built for
the real world.

Agentic AI for silicon engineering

• Engineer agentic LLM workflows across physical design (PD), design for test (DFT), and memory built-in self-test (MBIST) in Cisco Silicon One chip engineering.
• Build retrieval-augmented generation (RAG) systems for domain-specific retrieval and LLM-assisted engineering workflows.
• Fine-tune and adapt LLMs for specialized silicon engineering tasks.
• Develop and maintain AI infrastructure supporting model serving, RAG systems, and agentic workloads.
• Handle production AI operations, including deployment, monitoring, troubleshooting, and reliability through AIOps practices.

Large-scale iGaming technology company

• Built AI automation and agentic systems for product, marketing, and QA teams, converting manual analysis into structured, AI-assisted processes.
• Developed an AI-powered product marketing analyzer and competitor intelligence systems tracking odds, markets, and game placement across operators.
• Built an agentic QA system (Playwright) and a knowledge-base automation system that captures and structures internal domain knowledge.

Large-scale beauty & health e-commerce

• Architected and shipped a multi-agent reinforcement learning pricing engine across 24,000+ SKUs from scratch.
• Reduced PDP content production time by 60% for 10,000+ product listings with an end-to-end AI content system.
• Built a natural-language-to-SQL BI platform for 10+ non-technical stakeholders.
• Owned Airflow, MLflow, GitLab CI/CD, Docker/Kubernetes, Terraform, and AWS release workflows.

AI research and solutions organization

• Cut internal knowledge retrieval turnaround by 35% across a 200+ person organization.
• Designed domain-driven microservices for data-intensive AI workloads across 3 engineering teams.
• Implemented MLflow and Apache Airflow lifecycle patterns for training, deployment, monitoring, and rollback.

FMCG and retail operations

• Improved inventory workflow efficiency by 30% and cut overstock incidents by 20%.
• Increased customer engagement by 15% and marketing effectiveness by 25%.
• Aligned 3 cross-functional teams on a prioritized backlog and accelerated delivery by 2 sprint cycles.

AI and data tools for e-commerce sellers

• Built analytics pipelines over 1M+ data points and reduced weekly reporting from 8 hours to under 2 hours.
• Increased product catalog coverage by 3x and reduced data acquisition costs by 35%.
• Boosted system throughput by 40% and reduced query latency from 900ms to 300ms.

Technology and education holding group

• Built the institution's first centralized student data dashboard, cutting administrative processing time by 50%.
• Mentored 30+ junior developers through structured programming courses.