Skip to content
View karmendra8386's full-sized avatar

Block or report karmendra8386

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
karmendra8386/README.md

Karmendra Pandey

Practice Architect in the AI & ML practice at TEKsystems Global Services — I work with customers to enable and support their AI and ML platforms. I build open-source tooling for production AI agents: what an agent did, what it cost, and whether it should have been allowed.

Recognition

DevOps Dozen 2026 Finalist — three categories (public vote open through Dec 31, 2026):

DevOps Dozen 2026 Finalist — DevOps Advocate of the Year DevOps Dozen 2026 Finalist — Best End-User Application of AI in Software Delivery DevOps Dozen 2026 Finalist — Best DevOps Community Program or Event

Also: HW Foundation 50 Under 50 in AI — 2026 Honoree.

Selected work

Pinned Loading

  1. argus argus Public

    Argus — the security layer for AI models. An OpenAI-compatible LLM gateway that enforces security policy on every request and response: secrets blocked, PII redacted, prompt injection flagged, ever…

    Python

  2. agentic-ai-token-economics agentic-ai-token-economics Public

    Cost accounting for LLM agents: price any agent run — tokens, dollars, and cost per solved task — across models and frameworks.

    Python

  3. agentic-ai-aws-reference agentic-ai-aws-reference Public

    Use this: `Production reference architecture for AI agents on AWS: per-agent IAM identity, eval-gated model lifecycle, behavioral regression testing, data obfuscation, governance`

    Python

  4. mcplens mcplens Public

    MCP server security scanner with runtime behavioral verification — canary probes check that tools actually do what their descriptions claim, plus static rules and rug-pull diffing.

    Python

  5. agent-cost-mcp-server agent-cost-mcp-server Public

    MCP server for estimating, budgeting, and comparing LLM spend per task — reference implementation for awslabs/mcp RFC #4772, built on tokenecon.

    Python 1

  6. persistbench-bench persistbench-bench Public

    PersistBench: a benchmark for belief persistence and false-claim formation in LLM agents

    Python