AI engineer — agent reliability, LLM evaluation & guardrails. I check whether AI agents actually did what they said. Founder of Nikxius.
-
Nikxius
- https://nikxius.com
Pinned Loading
-
codex-council
codex-council PublicA four-agent Codex skill for evidence-based decisions: advocate, critic, value analyst, and judge.
Python
-
evalharness
evalharness PublicA small refuse-or-cite evaluation harness for LLM output. Standard library and pytest, no API keys, no network calls.
Python
-
receipts-collector
receipts-collector PublicCheck every "done" your coding agent said against its own logs. Local, standard library only, hashed output. Claude Code and Codex.
Python
-
sparespan
sparespan PublicGoverned critical-spare procurement with source-backed context, durable human approval, and controlled simulated execution. By Nikhil Sood.
Python
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.


