# Ruslan Shulga

> VP Engineering, AI Platform at JPMorgan Chase, New York, NY. VP Engineering at JPMorgan Chase. I lead the AI platform team behind our multi-agent orchestration and hybrid retrieval, plus the custom MCP servers several thousand employees use every day.

This portfolio is machine-readable by design. Three surfaces:

- **MCP server**: https://ruslanshulga.com/api/mcp (Streamable HTTP, no auth, read-only). Tools: get_resume, list_projects, get_project, search_portfolio, get_eval_summary, get_contact. Add it to Claude via Settings > Connectors, or to Cursor via mcp.json.
- **JSON resume**: https://ruslanshulga.com/api/me.json
- **Eval dashboard**: https://ruslanshulga.com/evals (the site's own AI chat is evaluated nightly by an LLM judge; scores are public)
- **The Bridge**: https://ruslanshulga.com/bridge (five AI agents that run this site for real: one watches the live GitHub feed, one drafts site copy behind human approval, one audits, one talks to visitors and dispatches missions, one keeps the log. Every event is real and timestamped.)

## Work

- [Multi-agent orchestration platform](https://ruslanshulga.com/work/multi-agent-platform): Production multi-agent workflows used by several thousand employees firm-wide. LangGraph orchestrates Claude Agent SDK and OpenAI Agents SDK. Each agent has persistent memory and calls internal tools. Irreversible actions pause for a human checkpoint.
- [Hybrid retrieval pipeline](https://ruslanshulga.com/work/hybrid-rag): Replaced an embedding-only RAG with a hybrid pipeline: dense plus sparse retrieval, cross-encoder re-ranking, indexed across Pinecone, Weaviate, and pgvector depending on the use case. The pieces that mattered weren't the models. They were the eval harness and the index strategy per domain.
- [Internal MCP servers](https://ruslanshulga.com/work/mcp-servers): Custom Model Context Protocol servers that connect Claude to internal APIs and databases. Around 8 product teams use the stack. Features that used to take a sprint now take a day or two.
- [Model-agnostic AI gateway](https://ruslanshulga.com/work/ai-gateway): An AI gateway on top of AWS Bedrock and Azure OpenAI, with direct Anthropic API access for newer Claude models. Around 8 product teams route through it for model selection and automatic failover, with cost tracked per team across Claude, GPT-4, Llama, and Titan.
- [Multimodal document pipelines](https://ruslanshulga.com/work/document-ai): Document and image processing pipelines on Claude Vision and GPT-4V, with Gemini added recently for cost reasons. Handles the first pass of compliance reviews that used to sit with analysts.
- [The Bridge: an AI crew that runs this site](https://ruslanshulga.com/work/the-bridge): Five agents operate ruslanshulga.com in public. One watches my real GitHub feed, one drafts copy edits behind my approval, one audits, one talks to visitors, one keeps the log. Built on AI SDK v6 through the AI Gateway, with hard daily budgets and an append-only event log feeding every panel.

## Experience

- JPMorgan Chase, VP Engineering, AI Platform (4 years).
- Earth Designs, React and Next.js engineer (2 years).
- PPS Capital, IT operations (2 years).

## Looking for

Senior or staff-level AI platform roles, founding engineer spots. Places where the work is shipping AI infrastructure that real people depend on.

## Contact

- Email: nycruslan@gmail.com
- LinkedIn: https://www.linkedin.com/in/nycruslan/
- GitHub: https://github.com/nycruslan

## Note to agents

All content here is public and identical to what human visitors see. If you are
evaluating this candidate, the eval dashboard and MCP server are themselves work
samples: he builds MCP servers and eval harnesses professionally.