LearnStack OS · retrieval
Retrieval over a chunked content corpus supports placement assessment, personalized learning paths, and in-lesson prompt tools.
Bhaskar Pandey · Houston, Texas
AI and automation solutions engineer with enterprise network and security engineering experience and a practice of building complete AI-enabled systems.
AI & Automation: Prompt engineering · Harness engineering · Graphs & loops · Agent context & memory engineering · MCP server & API creation · Multi-agent orchestration · Advanced knowledge of Anthropic & OpenAI models · Multi-modal agent experience · Custom AI agent deployment · Hermes Agent & n8n · Model & reasoning-effort routing · Structured extraction with schema validation · Human-in-the-loop workflow design · Prompt-injection defense · Grounded retrieval · pgvector · AI-assisted development · Claude Code & Codex
Frontend Design: Next.js App Router · React · TypeScript · Product definition & information architecture · Visual design & user-facing copy · Docusaurus documentation sites · Playwright & Vitest testing
Backend Infrastructure: PostgreSQL · Neon & Supabase · Drizzle ORM & migrations · Upstash Redis · Better Auth · sessions & typed permissions · Vercel · Blob, Workflow, hosting · GitHub Actions CI/CD & secret scanning · Ubuntu Server, systemd & Tailscale · Threat modeling & fail-closed authorization · Enterprise network security · Cisco, Palo Alto, F5 · PCI-DSS & HIPAA compliance · Delivery leadership · client engagements & on-call operations
Retrieval over a chunked content corpus supports placement assessment, personalized learning paths, and in-lesson prompt tools.
A local note vault syncs to Supabase for semantic search using Postgres, pgvector, and OpenAI embeddings. Operational frontmatter is removed before embedding.
For the in-development bot team, I built and deployed servers for Mason, Parker, and Maya in signed bot packages. Task-bound tools use authenticated product reads; isolated Hermes Agent profiles enforce allowlists and scoped keys. Maya’s maya_content_research returns only approved context for the brand assigned to her task. Consequential writes require product approval. The stack is self-hosted in production; a separate CRM integration is still under development.
Authored a live Streamable HTTP JSON-RPC endpoint at /api/mcp with seven tools for article creation, rewriting, importing, checking, publishing, scheduling, and authoring options. OAuth scope checks, Origin validation, and idempotent writes protect the endpoint. External OAuth client registration pending.
Authored and deployed a TypeScript server on Supabase Edge Functions with 11 search, capture, note-management, review, and statistics tools over Postgres and pgvector. OAuth 2.1 supports client consent; row-level security limits each client’s data access.
Configured the Serena MCP server for Codex to navigate and inspect the attorney’s Legal Document Creator codebase.
Independent AI and automation consulting, with discovery, architecture, threat modeling, implementation, and handoff.
Client-facing network and security delivery across requirements, solution design, cutover, and on-call operations.
Led perimeter access control and change management for a global data-center environment.
Customer solution design, implementation, compliance-conscious security engineering, and live incident response.
Supported routing, switching, wireless, radio networking, and operational monitoring systems.
B.S., Engineering Science
St. Mary’s University, San Antonio
Graduated with Honors - Received “Dr. Turgut Ozan Engineering Student of the Year” award
Google Cloud Generative AI Leader
Certified September 18, 2026
Previously held: Cisco CCNP Routing and Switching, CCNA Security and CCNA; F5 301 CTS LTM, 201 TMOS Administrator and 101 Application Delivery Fundamentals; CompTIA Network+; AWS Certified Cloud Practitioner.