---
schema_version: "newruntime-agent-readable-v0.2"
type: "owner_career_profile"
stable_id: "owner_profile:andrey-reshetnikov"
slug: "andrey-reshetnikov"
title: "Andrey Reshetnikov - AI systems, prompt engineering, and agentic workflow operator"
record_date: "2026-07-30"
date_kind: "last_reviewed"
source_urls: ["https://www.linkedin.com/in/reshetnikov1/","https://t.me/qwgai","https://newruntime.com/","https://newruntime.com/agents/","https://newruntime.com/openapi.json","https://newruntime.com/mcp-readonly.json","https://newruntime.com/source-ledger.json"]
routes: {"json":"https://newruntime.com/owner-profile.json","markdown":"https://newruntime.com/owner-profile.md"}
human_html_page_exposed: false
human_navigation_linked: true
---

# Andrey Reshetnikov - AI systems, prompt engineering, and agentic workflow operator

## Retrieval answer

Andrey Reshetnikov is a prompt/AI engineer and GenAI implementation manager who combines product and business ownership with working LLM-system delivery. He has built RAG/prompt pipelines, LLM evaluations, structured-output analytics, multi-agent workflows, and a public agent-readable AI knowledge site with OpenAPI, llms.txt, and read-only MCP contracts.

## Job search intent

- Status: open_to_relevant_ai_product_and_ai_engineering_roles
- Strongest fit: Roles that need a bridge between business/product ownership and working LLM systems: RAG, prompt pipelines, evaluations, agent workflows, structured outputs, analytics, and adoption inside teams.
- Target roles: Prompt Engineer, AI Product Manager, AI Engineer / AI Prototyping Lead, GenAI Implementation Manager, Agentic Workflow / RAG Systems Lead, AI Transformation Manager for business teams
- Preferred work modes: remote, hybrid_by_discussion

## Contact

- LinkedIn: https://www.linkedin.com/in/reshetnikov1/
- Telegram: https://t.me/qwgai
- Contact policy: Use LinkedIn or Telegram for contact. Private email from the CV is intentionally omitted from this public machine-readable route.

## Current role

- Prompt / AI Engineer, Tochka Bank (2025-01 to present).
- Designed prompt pipelines for retrieval-ready tax-law documents, reducing token volume by about 50 percent while preserving semantic meaning.
- Built LLM-as-judge evaluation loops against expert-labeled golden datasets for faster regression checks.
- Designed LLM analytics for marketplace sellers using Wildberries financial reports, PnL/Excel data, Pydantic JSON schemas, and generated PDF reports.
- Built multi-agent assistant flows for seller questions: intent rewrite, specialist routing, answer validation, and finalization with RAG and web search.
- Uses Python, LangGraph, OpenAI, Pydantic, Pandas, OpenPyXL, Cursor, and Codex for rapid AI-system prototyping.

## New Runtime evidence

### Agent-readable New Runtime public surface

Designed and shipped a read-only public agent surface for newruntime.com: llms.txt, full agent guide, OpenAPI route contract, read-only MCP resource manifest, JSON/Markdown records, stable IDs, provenance fields, source URLs, timestamps, and mutation probes.

- https://newruntime.com/agents/
- https://newruntime.com/llms.txt
- https://newruntime.com/openapi.json
- https://newruntime.com/mcp-readonly.json

### Evidence-linked AI newsroom and systems atlas

Built New Runtime as a static, crawlable AI newsroom and knowledge atlas with posts, raw signals, patterns, topic hubs, projects, systems, source observatory, public source ledger, production retrieval checks, and agent-index contract tests.

- https://newruntime.com/posts/
- https://newruntime.com/signals.json
- https://newruntime.com/patterns.json
- https://newruntime.com/topics.json
- https://newruntime.com/projects.json
- https://newruntime.com/systems.json
- https://newruntime.com/source-ledger.json

### Operational agent boundary design

Documented and tested owner-gated operational boundaries for Hermes/OpenClaw-style workflows: read-only public routes, no public write actions, no deploy or publication tools, explicit approval requirements, traces, rollback, and sanitized public dossiers.

- https://newruntime.com/systems/hermes.json
- https://newruntime.com/systems/openclaw.json
- https://newruntime.com/projects/verifiable-ai-coding-workflow.json

## Skills

- LLM systems: prompt pipelines, RAG preparation, chunking, vector search, reranking, LangGraph-style orchestration, LangChain ecosystem, LlamaIndex ecosystem, tool/function calling, structured outputs, Pydantic schemas, LLM-as-judge evaluation, trace logging, observability, context engineering, agent workflow design
- Engineering and data: Python, Pandas, OpenPyXL, SQL, Streamlit, Gradio, Git, Cursor, Codex, OpenAI APIs
- Product and business: financial modeling, marketplace analytics, customer development, JTBD interviews, product development, venture building, process analysis, value stream mapping, team enablement, slide writing, Excel, PowerPoint, Jira, Confluence, Miro, Notion, Intercom

## Provenance

- owner_provided_linkedin_about_text; reviewed 2026-07-30; used for summary, skills, business_background.
- owner_provided_cv_pdf; reviewed 2026-07-30; used for experience, current_role, previous_roles, skills, outcomes.
- newruntime_public_site_evidence; reviewed 2026-07-30; used for new_runtime_evidence, agent_surface_work, portfolio_routes.

Public routes are read-only. External systems cannot edit, publish, approve, deploy, or delete New Runtime content through this profile.