Who am i?

i'm an ai engineer who builds and ships ai systems end to end: architecture, implementation, and the unglamorous part where it has to keep running in production.

my toolkit is python, langgraph, crewai, langchain, fastapi and docker. i care about jwt rbac, rate-limit handling, load balancing and observability as much as prompt design and model accuracy, because that's usually what separates a demo from something people can actually depend on.

i founded agentnine, an independent directory of open-source ai agents. finding an agent is easy; finding out how to set it up and what access it wants is not. every listing shows the repository, install steps for windows, macos and linux, what the agent can reach, and what's still undocumented. i lead product direction and technical decisions, working with a small team across product, engineering and launch.

before that, as an ai/ml trainee at technosavvys, i fine-tuned a bert classifier on 500+ student reviews with hugging face. it hit 88% accuracy, a 15% lift over a tf-idf baseline, and i shipped the streamlit dashboard mentors actually used.

i'm open to ai engineer and applied ai roles, remote or hybrid.