AI and automation
RAG, LLM applications, agents, MCP, semantic search, knowledge ingestion, evidence grounding, transformers, and fine-tuning.
Designed and built evidence-backed support intelligence for financial-services engineering, then evolved it into an MCP-first, skill-driven runtime.
50+
users since launch
~1.4K
estimated queries since June 2025
~650
estimated conversations since June 2025
545
measured MCP calls in the latest 3 weeks
The conversation and query totals use a linear adoption estimate. At four MCP calls per query, the same model suggests approximately 5,000-6,000 cumulative data-tool calls.
- Built natural-language investigation across defect metadata, comments, documentation, and operational knowledge while preserving the evidence behind each answer.
- Owned the software development lifecycle across ingestion, user management, documentation, production delivery, and proactive weekly features.
- Expanded horizontally to additional financial-services products and vertically through product-specific capabilities.
- Worked across tenant lifecycle, identity, secrets, artifacts, Docker, Kubernetes, Jenkins, logging, monitoring, and upgrades.
- Supported Corporate Security and Assurance Process (CSSAP) reviews and internal production approvals.
- Built root-cause workflows, reusable skills, plugin delivery, and repository-to-wiki automation with controlled updates.