Karmendra Pandey
karmendra8386
·
AI & ML interests
Practice Architect in the AI & ML practice at TEKsystems. I build open tools and benchmarks for AI agent cost and governance — tokenecon, Argus, and the Cost Per Correct benchmark.
Recent Activity
new activity 4 days ago
gaia-benchmark/leaderboard:Show best submissions per team, agent and model family by default updated a dataset 8 days ago
karmendra8386/cost-per-correct new activity 10 days ago
mario0369/llm-cost-same-prompt:Are failed runs in the per-call averages, or only in probe_runs.csv?Organizations
None yet
Show best submissions per team, agent and model family by default
1
#113 opened about 1 month ago
by 
zsewasdsdsd
Are failed runs in the per-call averages, or only in probe_runs.csv?
#2 opened 10 days ago
by karmendra8386