# Pavel Borisov — AI Engineer

> 20+ years designing high-load systems and platforms for AI products. Founder and ex-CTO of the AI startup SelfieBot. I specialize in production-ready infrastructure for local LLM/VLM inference, RAG systems, and MLOps automation — across Cloud and Bare Metal.

## About

**Core AI Engineer**

It all started with Assembly language for the ZX Spectrum: one day I wrote my own loader so that games loaded from cassette almost twice as fast as with the standard one. That was when I first realized I was more interested in understanding how a system works than simply using it.
Later, working as a DBA, I squeezed maximum performance out of SQL where others threw up their hands. Little has changed since then — I’m still more interested in building complex systems, finding bottlenecks, and making them faster, more reliable, and easier to operate.
Today, that approach has led me to AI Infrastructure. I design production-ready platforms for LLM/VLM, build RAG architectures, automate MLOps, and deploy local inference — from bare metal to the cloud. I enjoy working at the intersection of infrastructure, data, and applied AI, where every engineering decision can be measured by performance, cost, and reliability.
My experience in photography, design, and art direction added one more principle: a good system should not only be fast inside, but pleasant on the outside too. That’s why I pay equal attention to backend architecture, interfaces, and user experience.

### Expertise
- **AI / ML Infrastructure** — Deploying and tuning local inference (vLLM, NVIDIA Container Toolkit, CUDA); Building RAG architectures (Qdrant, pgvector, LangChain / LangGraph, LangFuse); Computer Vision pipelines (OpenCV, PaddleOCR, DLib)
- **Orchestration & DevOps** — Fault-tolerant clusters (Kubernetes, Docker); IaC (Ansible), CI/CD, observability stacks (Grafana, Prometheus); Zero-Downtime migrations and High Availability architectures
- **Backend & Data** — High-throughput services in Go and Python (Asyncio, FastAPI, uv); Distributed systems (NATS), PostgreSQL on steroids; Data architecture, ETL/ELT, observability
- **Product Mindset** — Team leadership and zero-to-launch AI product experience (as CTO); Business metrics, data architecture, and design processes; Translating business pain into technical requirements

### Interests

AI Infrastructure, Computer Vision, LLM Agents, Web Design, Digital Storytelling, photo & audio gear.

## Career History

### AI Infrastructure & Highload Systems — Lead Engineer

Period: March 2019 — Present (Consulting / Outsourcing)

- Designing, scaling, and migrating infrastructure for high load and AI/ML services for clients in e‑commerce, AdTech, and IoT.
- Deploying and containerizing AI solutions: computer-vision pipelines, local LLM inference, and RAG systems.
- Building fault-tolerant architectures (Kubernetes, Docker) and CI/CD automation to reduce Time-to-Market.
- Tuning PostgreSQL performance for high-throughput analytical workloads.
- Owning the full engineering lifecycle: business requirements → Cloud / Bare Metal release.

### [АТТЭК (ATTEK)](https://www.centrattek.ru) — Lecturer / Mentor — AI, QA, Backend

Period: October 2025 — December 2025 (Saint Petersburg)

- Designed and ran a practical course on Python, SQL, automation QA, and modern AI workflows.
- Brought 26 pre-junior students up to production-ready level.
- Supervised real student projects using LLMs and Cloud PostgreSQL.
- Promoted best practices for backend dev, testing, and system design.
- Acted as mentor and tech lead on capstone projects — practical engineering over dry theory.

### [Selfiebot.ru](https://selfiebot.ru) — Founder

Period: November 2023 — August 2025 (Moscow)

- Built from zero to first revenue an AI-powered lead-gen and face-recognition platform.
- Designed and shipped a REST API (FastAPI) and CV pipelines (DLib, OpenCV).
- Integrated LLMs (GPT and local), driving marketing-conversion gains.
- Built the frontend (React + Tailwind) and infrastructure (Docker, Debian, monitoring stack).
- Cut downtime to <1% via observability (Prometheus, Grafana).
- Reached first revenue and ~1,000 active users.
- End-to-end ownership: product, architecture, infrastructure, and team processes.

### [Raskachaem.ru](https://raskachaem.ru) — Regional Director

Period: March 2014 — March 2017 (Arkhangelsk)

- Ran regional operations in a competitive media market.
- Led a distributed team of ~10 photographers and content makers.
- Coordinated intensive event coverage on tight deadlines.
- Built partnerships with venue owners and media outlets.
- Negotiated commercial partnerships and ad campaigns.

### [PJSC «SIA International-Arkhangelsk»](https://ru.wikipedia.org/wiki/%D0%A1%D0%98%D0%90_%D0%93%D1%80%D1%83%D0%BF%D0%BF) — Head of IT Department, DBA

Period: November 2003 — December 2015 (Arkhangelsk)

- Managed IT infrastructure and data architecture for a large pharma distributor (central hub + 2 branches, 200+ corporate clients). 6-person team.
- Zero-Downtime migration: moved the data center and the company to a new building without halting operations or shipments.
- High Availability: designed and scaled a distributed, fault-tolerant order-processing system across the branch network.
- Migrated the core ERP system with no data loss and no tech downtime.
- Delivered monitoring, backup, and electronic-document workflows for the branch network and clients.
- Built bespoke wholesale-trading software (Python, Bash, SQL).
- Full management loop: IT budgets, risk, hiring, onboarding, cross-functional alignment.

### [City of Arkhangelsk Administration](https://arhcity.ru) — Lead Engineer

Period: January 2002 — November 2003 (Arkhangelsk)

- Designed and supported LAN networks for administrative departments.
- Worked with internal units to deploy IT solutions.
- Technical support and infrastructure problem-solving under strict regulations.

## Core Competencies

### AI / ML & Agents

`AI`, `Agents`, `RAG`, `LangChain`, `LangGraph`, `LangFuse`, `Qdrant`, `pgvector`, `PaddleOCR`, `OpenCV`, `vLLM`, `NVIDIA Container Toolkit`

### Languages & Runtime

`Python`, `Golang`, `Bash`, `Asyncio`, `FastAPI`, `n8n`

### DevOps & Infra

`DevOps`, `Linux`, `Docker`, `Kubernetes`, `Ansible`, `Nginx`, `Git`, `Prometheus`, `Grafana`, `CI/CD`

### Data & Storage

`PostgreSQL`, `MariaDB`, `MSSQL`, `Firebird`, `SQLite`

## Public Repositories

### hermem

URL: https://github.com/pavelveter/hermem
Stack: Go, SQLite, MCP, OpenAPI, SDK

Ultra-fast graph memory layer (GraphRAG) for AI agents — solves long-term associative memory with minimal latency.

### research-agents-swarm

URL: https://github.com/pavelveter/research-agents-swarm
Stack: LangGraph, LangFuse, Qdrant, Ollama

Autonomous 5-agent swarm with the LLM-as-a-judge pattern for deep internet research and report generation.

### llm-gateway

URL: https://github.com/pavelveter/llm-gateway
Stack: Python, LLM, Proxy

Lightweight proxy server unifying B2B inference: a single endpoint over many LLM backends with automatic failover.

### recognition

URL: https://github.com/pavelveter/recognition
Stack: Python, uv, OpenCV, DLib

Face recognition & matching module (Face ID / Computer Vision) powering SelfieBot — matches selfies of guests against event photo reports. Tuned for high-throughput batch processing.

### regognition

URL: https://github.com/pavelveter/regognition
Stack: Go, Computer Vision, RetinaFace, ONNX

Go rewrite of the face matching module: faster and more accurate (512-dim vectors vs 128-dim), with lower memory footprint for local workloads.

## Commercial Projects

### Enterprise RAG for engineering documentation

Tagline: Intelligent search across large-scale technical documentation (PDFs, scans, photos, tables)

Build intelligent search across a large corpus of technical documentation — PDFs, photos, scans, tables and documents in many formats.

**Solution** — Designed and shipped an Enterprise RAG platform with sophisticated document parsing.

**Implemented** — semantic document chunking; Parent / Child retrieval; scalable OCR pipeline; GPU inference; asynchronous document processing; auto-scaling OCR workers

**Results:**
- natural-language search across technical documentation
- operates on tens of thousands of documentation pages
- drastically reduces manual information lookup for engineers
- architecture fit for industrial workloads and scaling

**Stack:** PaddleOCR, Table Transformer, LayoutLM, pdfplumber, openpyxl, python-docx, LlamaIndex, pgvector, NATS JetStream

### Bastiondom

Tagline: AI-driven lead-generation platform for a country-house developer

**AI Infrastructure** — Smart Telegram bot powered by large language models; Natural-language dialogues with prospects; Lead qualification and interactive quizzes

**Referral System** — Two-tier referral program inside the bot; Built to scale viral traffic

**Growth Hacking** — Drop-domain scheme: custom redirect architecture on high-traffic drop domains

**Results:**
- Drop-domain scheme + AI lead processing lifted the weekly qualified-lead flow from dozens to thousands

**Stack:** LLM, Telegram Bot API, Drop-domain redirects, Referral mechanics

### Teonaperfume.ru

Tagline: Premium niche-perfume e‑commerce — technically optimized
Live URL: https://teonaperfume.ru

A high-performance premium-segment e‑commerce project with a deep focus on technical optimization and virality. Minimal luxury design paired with extremely fast page loads.

**Tech & Features** — Custom seamless i18n; Dynamic React Router for instant catalog navigation without page reloads; Dynamic QR codes per product for instant sharing

**SEO Architecture** — Dedicated SEO layer for product cards; Outranked competitors in SERP via structured data and semantics

**Results:**
- Organic lead flow tripled vs. the first version
- Site paid itself back in week one post-launch from new organic sales alone

**Stack:** React, Dynamic Routing, i18n, Dynamic QR-codes, SEO schemas, Core Web Vitals

### Teambelki.ru

Tagline: Interactive multi-landing for an event agency
Live URL: https://teambelki.ru

A heavy-media platform built to differentiate the brand in the corporate-events market from its competitors.

**Tech & Features** — Ultra-fast heavy-content loading via mobile-first + wow animations; Lightweight PocketBase backend & admin panel — client manages content without server bloat

**Bypass Restrictions** — Custom video player & streaming through Rutube, fully isolated from external throttling and access restrictions

**AI Optimization** — Content & metadata tuned for the new Google AI-search algorithms (SGE / Gemini Search); Dominance in next-gen search results

**Results:**
- Platform differentiates the brand from competitors in the corporate-events segment
- Loading speed and SEO layer for Google AI search yield durable competitive advantage

**Stack:** Mobile-First, PocketBase, Custom Rutube player, SGE / Gemini SEO

### Dvevilochki.ru

Tagline: Premium catering landing
Live URL: https://dvevilochki.ru

A case study in how a product approach + modern tech can rescue marketing even when there are no quality source materials from the client. The final visual looks premium and makes the food sing.

**Tech & Features** — Full mobile-first UI with a strong focus on visual content; Main problem — low-res, poor-quality dish photos from the client — solved at the production stage

**AI Generation** — Generative AI applied for deep restyling, food design and image upscale; Menu photos reworked to restaurant-grade quality

**Results:**
- The premium visuals turned the site into a powerful sales tool that earns trust from second zero

**Stack:** Mobile-First, Generative AI, Image upscale, Food design pipeline

## Media & Public Talks

- [“AI Agents Demand Money and Criticize People” — expertise in software development with AI agents; article edited with Hermes Agent](https://www.rosbalt.ru/news/2026-07-22/ii-agenty-trebuyut-dengi-i-kritikuyut-lyudey-neyroseti-vyshli-iz-pod-kontrolya-5643115)
- [“When a Neural Network Becomes a Prophet: How AI Creates New Religions” — a story of how a chatbot can convincingly play the role of a digital prophet](https://www.rosbalt.ru/news/2026-08-18/kogda-neyroset-stanovitsya-prorokom-kak-ii-sozdaet-novye-religii-5662212)
- [“The World Wide Web Is Tearing: Who Will Lead the Internet Restrictions Race?” — analysis built with research-agents-swarm](https://www.rosbalt.ru/news/2026-06-17/vsemirnaya-pautina-rvetsya-kto-budet-pervym-v-gonke-ogranicheniy-interneta-5619135)
- [“A New Data Center Will Appear in St. Petersburg: Why It’s Needed” — infrastructure & power footprint analysis](https://www.rosbalt.ru/news/2026-04-08/novyy-tsod-poyavitsya-v-peterburge-zachem-on-nuzhen-5577204)
- [“Neural Networks Deceive People, AI Creates Digital Hallucinations” — LLM internal mechanics & token probability analysis](https://www.rosbalt.ru/news/2025-08-14/neyroseti-obmanyvayut-lyudey-a-ii-sozdaet-tsifrovye-gallyutsinatsii-kto-za-etim-stoit-5456186)

## Recognition

- **Silver medalist — Programming Olympiad** — Verified academic track record in Computer Science.
- **Author and speaker on AI and technology** — Regular subject-matter expert in federal media (incl. IA Rosbalt, 8M unique monthly visitors) on AI tech and IT infrastructure.
- **Workshops on communication & chaos management** — Designed and delivered practical workshops on communication, negotiation, and collaboration in high-pressure, uncertain, and conflict-prone environments.

## Languages

- Russian — Native
- English — B2

## Education

- **2002** — Northern (Arctic) Federal University named after M.V. Lomonosov, Institute of Information and Space Technologies, Computer-Aided Design Systems (Higher education)

## Contact

- Email: a@pavelveter.com
- Telegram: https://t.me/pavelveter (@pavelveter)
- GitHub: https://github.com/pavelveter
- Citizenship: Russia, St. Petersburg

## For AI agents
- [llms.txt](https://pavelveter.com/llms.txt): short index of this site
- [llms-full.txt](https://pavelveter.com/llms-full.txt): full résumé as Markdown
- [Sitemap](https://pavelveter.com/sitemap.xml): all human-readable pages

_Last updated: July 21, 2026._