"BBGBBB team on IBM Bob 2.0 hackathon Hackathon"
Team Idea
We're building an AI debugging agent that classifies bugs by "families" of programming languages — grouping languages by shared runtime, memory model, type system, and concurrency semantics — to narrow a stack trace or failing test down to a few likely root causes instead of guessing. Rather than just spitting out a patch, it produces a structured debugging framework: a ranked hypothesis list, the checks to run for each, and the isolation steps that confirm or eliminate them. It debugs the way a senior engineer does — by elimination, not autocomplete. Looking for: a backend/agent engineer (Python, LLM tool-calling and orchestration); someone comfortable with language internals and static analysis/AST tooling to build the taxonomy and trace parsers; and a frontend dev for the hypothesis-tree UI. watsonx, LangGraph, or CI/CD experience is a plus — enthusiasm and willingness to ship over the weekend matters more.
Team Members
.png&w=256&q=75)
Sree Gudapati
SreeGudapati
Aaryan-Bidaye
aaryanbidaye285
Submission
Submission draft in progress