Skip to content
Bharat Khanna Bharat Khanna

AI Engineer Portfolio

Building dependable AI systems that ship.

From agent design to eval rigor, I build AI systems that hold up in production.

Agents RAG EvalOps LLM Infra
Explore Projects Read AI Articles

6

Knowledge Articles

3

Production Projects

5

External Publications

4

Connected Content Rails

What I Build for AI Teams

Agent Architecture and Orchestration

Multi-step workflows, tool routing, memory contracts, and execution reliability for complex AI tasks.

RAG and Retrieval Systems

Robust ingestion, vector retrieval quality, and context selection strategies for production-grade answer accuracy.

EvalOps, Guardrails, and Observability

Evaluation harnesses, regression checks, runtime policy controls, tracing, and measurable quality loops.

One ecosystem. Four content rails.

Your latest thinking and builds, linked in one place so readers can move from post to code to publication without friction.

Recent Writing

April 11, 2026

Guardrails as Runtime Policy in Multi-Agent Systems

A systems-oriented look at why guardrails belong inside orchestration, with benchmark results from agentic-runtime across support triage and research-and-retrieval workloads including adversarial cases.
agents langgraph python guardrails
March 8, 2026

Skills vs MCP: Designing the Next Generation of AI Agent Architecture

Not all problems an AI agent solves are the same. This piece unpacks the distinction between Skills (knowledge layer) and MCP (execution layer) — and why separating them leads to more efficient, maintainable agent systems.
mcp agent-skills context-engineering ai-architecture

Collaboration

Building AI capabilities that survive production.

Open to conversations around agent platforms, RAG architecture, evaluation systems, and reliability-first AI engineering.

Start a Conversation Connect on LinkedIn