Skip to content
View husky-parul's full-sized avatar
  • Red Hat
  • Netherlands
  • 13:26 (UTC +02:00)
  • X @piparul

Block or report husky-parul

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
husky-parul/README.md

Hi, I'm Parul Singh

Principal Software Engineer at Red Hat, Office of the CTO.

I work at the intersection of AI agent security, cloud-native infrastructure, and sustainability. My focus is on trust, identity, and observability for autonomous AI agents — building systems where machines can prove who they are, what they're allowed to do, and how much energy they burn doing it.

SIG Lead, CoSAI ADLC — leading the Agent Development Lifecycle security workstream within the Coalition for Secure AI (an OASIS Open project with members including Google, Microsoft, OpenAI, Anthropic, and others). ADLC extends traditional SDLC frameworks to address the unique challenges of autonomous decision-making, non-deterministic behavior, and runtime tool invocation in AI agent systems.


What I'm Building

  • trust-graph — Trust-aware dataplane for agentic systems — consistent authorization enforcement across multi-hop agent → tool → resource chains. Proposed in both CoSAI WS4 and kagenti
  • Kepler — Kubernetes energy observability using eBPF (CNCF Sandbox)
  • PEAKS — ML-driven power-efficiency-aware Kubernetes scheduler
  • modelcard-tools — OCI-native model card tooling for AI transparency

Recent Work

  • Trust Graph Dataplane — Building a trust-aware authorization dataplane for agentic systems with DAG-based delegation and AuthBroker RFC
  • Agent Registry — Experimenting with trust graph decision surfaces and zero-instrumentation agent lineage via OpenTelemetry
  • Llama Zero-Trust Stack — Added MCP Weather Tool with scope-based access control to the Llama zero-trust stack
  • Kepler Maintainership — Maintaining Kepler in the CNCF, including validation exports and power model improvements
  • PEAKS Benchmarks — Published performance benchmarks for power-efficiency-aware Kubernetes scheduling

Speaking

I've spoken at KubeCon + CloudNativeCon, PyTorch Conference, Open Source Summit, and Red Hat DevConf (US & CZ) on topics including energy observability, power-aware scheduling, AI model transparency, and secure agentic systems.

See my talks on Sessionize.


Other Things

  • Red Hat Patent Program Inventor
  • IBM Quantum Associate Ambassador
  • Contributor to CNCF Environmental Sustainability Working Group

LinkedIn · Sessionize

Pinned Loading

  1. trust-graph trust-graph Public

    Python

  2. trust-graph-dataplane trust-graph-dataplane Public

    HTML 1