Amrik Singh Khalsa is an AI engineer and software architect in Bangalore, India, building AI agent systems and the security gates around them at Okta (Auth0).
Answers come from Amrik's site and resume, matched on this page. This is not a live AI model.From this site's content. Not a live AI model.
Building AI agents and the frameworks, protocols and security gates that make them safe to run in production.
Measure before you claim.
Read the source, not the summary.
Retract in public.
Most of the systems I've built exist because a number was being quoted that nobody had verified.
Executable agent skills behind my CI gates
Upstream PRs merged at NVIDIA, Microsoft & Warp
PRs merged on AI platform & security at Okta
Architecture & research docs published
AI agent infrastructure and the security, identity and governance layers around it.
Led the team behind the core platform of India's first equity secondary market.
Led teams through growth while building shared infrastructure across product lines.
Mentored underprivileged students while studying for a B.Tech in Computer Science at Shri Vaishnav Vidyapeeth Vishwavidyalaya (SVVV), Indore, Madhya Pradesh.
Fixed tool-call linkage in fine-tuning trajectories and in both training backends.
Configurable secret redaction, decision evidence in a tamper-evident audit log, and a safer kill switch.
Executable shell scripts opened from a file:// URL run in the terminal instead of the editor.
Measure it before you claim it, and read the source rather than the summary.
A system that says what it couldn't reach beats one that confidently invents.
The best ideas come from fresh perspectives and new experiences.
Like biking uphill, persistence leads to amazing views.
Building solutions that empower people and create real change.
Every sunset, every trip, every challenge teaches something new.
Amrik Singh Khalsa is an AI agent systems and software architect based in Bangalore, India. He is a Senior Software Engineer at Okta (Auth0), where he builds multi-agent systems, MCP servers, an A2A protocol runtime and the CI security gates for executable AI agent skills. He has five years of full-stack engineering experience.
Yes. Amrik Singh Khalsa is an AI engineer who specializes in AI agents. At Okta (Auth0) he architected a multi-agent fleet that root-causes production escalations with cited evidence, delivered six agent-platform proofs of concept, and built a RAG support agent on pgvector that cut internal query resolution time by 40%.
That is Amrik Singh Khalsa, who publishes code on GitHub as amriksingh0786 and is on LinkedIn as amriksinghkhalsa. He works on AI agent infrastructure at Okta (Auth0) in Bangalore. Before that he was Founding Engineer and Tech Lead at Incentiv and Frontend Lead at Augmento Labs, both also in Bangalore.
Amrik Singh Khalsa studied Computer Science at Shri Vaishnav Vidyapeeth Vishwavidyalaya (SVVV), Indore, Madhya Pradesh, earning a Bachelor of Technology between 2017 and 2021. While studying he was President of the non-profit PRAYATNA and founded the CodeChef SVVV chapter. He has worked as a software engineer in Bangalore since 2021.
Amrik Singh Khalsa owns the CI security gates for an org-wide marketplace that distributes 170+ executable AI agent skills at Okta (Auth0). He shipped three merged gates: an LLM-backed security scanner, a static-analysis pre-commit hook and a duplicate-skill gate. GitHub OIDC-federated IAM roles supply short-lived model credentials, so no long-lived secrets are needed.
Yes. Amrik Singh Khalsa has delivered six agent-platform proofs of concept end to end at Okta (Auth0). They include an A2A protocol runtime with more than 1,200 tests passing, MCP servers, agent memory with retrieval grounding on AWS Bedrock AgentCore, and observability for agent context and cost.
Amrik Singh Khalsa has eight merged pull requests across three projects: the NVIDIA NeMo Agent Toolkit, the Microsoft Agent Governance Toolkit and the Warp terminal. They fix tool-call linkage in fine-tuning data, add configurable secret redaction and tamper-evident audit evidence for agents, and make executable scripts opened from file:// links run in the terminal. Together they added 104 unit tests.
Before Okta (Auth0), Amrik Singh Khalsa was Founding Engineer and Tech Lead at Incentiv from April 2024 to June 2025. The core platform his team delivered drove $5M+ in ESOP liquidity and onboarded 30+ startups and 200+ investors in six months. From 2021 to 2024 he grew from Software Engineer to Frontend Lead at Augmento Labs.
For agentic AI systems, Amrik Singh Khalsa works with the Claude Agent SDK, MCP, A2A, LangChain, LangGraph, RAG with pgvector and AWS Bedrock AgentCore. His platform stack covers TypeScript, Python, Go and Rust, with Next.js, NestJS, PostgreSQL, Terraform and GitHub OIDC across AWS, Azure and GCP.
Yes. Amrik Singh Khalsa takes on consulting, advisory and freelance engagements on AI agent architecture, multi-agent systems, MCP servers, RAG and AI agent security. Every engagement is scoped in writing first. To start, send a short brief through the Work with me page at amriksinghkhalsa.com/work-with-me, or email singhamrikkhalsa@gmail.com.
Yes. Amrik Singh Khalsa is open to remote and full-time roles as a senior engineer, software architect or tech lead on AI agent systems. He is based in Bangalore, India. His resume, covering five years of building and leading full-stack platforms and eight merged upstream pull requests, is at amriksinghkhalsa.com/resume.
Email Amrik Singh Khalsa at singhamrikkhalsa@gmail.com, or reach him on LinkedIn at linkedin.com/in/amriksinghkhalsa and on GitHub as amriksingh0786. He is based in Bangalore, India, and is open to conversations about AI agent systems, multi-agent architecture, MCP servers and AI agent security. For consulting or freelance projects, send a brief through the Work with me page.
