Making things happen

Automated core business workflows with agentic RAG pipelines, reducing manual processing time by 15%.

Architected client-specific vector-database stacks that powered real-time personalization across products, boosting CTR by 10%.

Engineered a deep-clustering segmentation model that improved audience-profiling relevance.

Built a next-best-action recommender (reinforcement-guided) that raised credit-line utilization by 7% in early-access.

Designed an automated evaluation pipeline (LangSmith, Weights & Biases) that cut model-benchmark cycle time by 40%.

Prototyped generative-AI use cases and briefed the C-suite, authored a company-wide AI roadmap.

Maintained ISO-42001-aligned ML-lifecycle documentation, resulting in zero audit findings.

Facilitated cross-functional alignment—frontend, backend, product, and domain SMEs—securing consensus on AI feature scope and acceptance criteria, which cut scope churn and kept key releases on schedule.

Spearheaded AI/ML use-case discovery and kicked off a data-infrastructure overhaul (AWS Glue, Delta Lake, Airflow)

Built an automated vendor-selection recommender that trimmed 10 min / relocation and saved ~$500 K per year.

Engineered a forecasting model that raised business-projection accuracy by 8 % driving downstream gains in staffing efficiency and customer satisfaction score.

Directed development of a client-expense prediction model, beating the legacy baseline by 14 %.

Launched a client-facing Gen-AI chatbot. Also, an internal HR and RFP assistant reducing processing time by 20%.

Led all AI initiatives, coordinating a six-person team (2 DS, 2 DE, 1 UI dev, 1 scrum master).

Improved fine grain classification accuracy across 13k SKUs from 76% to 90.5%

Revamped data augmentation & annotation of bot captured images

Led a team to develop fixed camera system as an alternative to robot mounted cameras

Pruned facial landmarking models for embedded applications