Open to opportunities

Akim Pavlov

@pavakim

Message

Applied AI/LLM engineer with 3+ years building production LLM systems and Python backend automations.

Russian Federation

Message

What I'm looking for

I’m looking for a role where I can own the full lifecycle of production LLM/RAG systems in Python—balancing quality, LLM cost, and scalability—while integrating with real business tools and maintaining independent control of architecture and logic.

I’m an Applied AI/LLM Engineer with 3+ years of experience developing Python-based software solutions and integrating AI tools into real-world business processes. I focus on production-oriented LLM systems—AI assistants, RAG architectures, customer communication automation, LLM routing and fallback, context management, moderation, and human-in-the-loop workflows.

I work end-to-end: translating business requirements into working AI systems, designing implementation architectures around client infrastructure, and owning backend development, third-party integrations, deployment, and optimization. Across projects for 5+ external client companies, I reduced average response time from about 5 minutes to 2–3 seconds and cut operator workload by 95%. I’ve also led omnichannel AI systems for Telegram, Jivo, and Envybox, including automated end-to-end social media management using generative AI.

Alongside delivery, I publish and build—author and co-author of 20+ research publications on AI/LLMs and related areas like speech intelligence and speaker diarization, plus software engineering. I bring a research-grade understanding of models while keeping a strict production mindset: I optimize for generation quality, LLM cost, scalability, and engineering decision control.

Experience

Work history, roles, and key accomplishments

Current

Founder / Lead AI Engineer

Current

Lekha LLC

Dec 2025 - Present (7 months)

Founded and led development of an AI product in the Student Startup program, building end-to-end social media management automation using generative AI. Designed and implemented an omnichannel AI system (Telegram, Jivo, Envybox) with a Pinecone-based RAG architecture backed by Google Sheets, reducing response time to 2–3 seconds and operator workload by 95%.

Applied AI Engineer / Backend Developer

Inevils LLC

May 2023 - Jan 2026 (2 years 8 months)

Delivered AI tools and business-process automation for 5+ client companies, independently designing implementation architectures around each client's constraints. Built production-oriented LLM systems including RAG, LLM routing/fallback, context management, moderation, and human-in-the-loop workflows, reducing average response time to 2–3 seconds and operator workload by 95%.

Programmer / Curriculum Developer

Algorithmics

Feb 2022 - Dec 2025 (3 years 10 months)

Developed and delivered an online learning platform using Django and JavaScript, creating educational programs and teaching hands-on technical projects. Implemented LLM integrations using OpenAI, Gemini, and YandexGPT, including routing/fallback, context management, request moderation, and automatic handoff to live operators.

Python Django JavaScript HTML CSS Lua Roblox Development Curriculum Development Online Learning OpenAI API Gemini API Fallbacks Dialogue History Context Managers Agent Handoff API Integration Programming Instruction

Software Developer / QA Tester

DMT LLC

Jun 2024 - Aug 2024 (2 months)

Deployed backend solutions using Docker, Nginx, Linux, and cloud/VPS environments and performed functional testing of hardware products and associated software. Identified defects and contributed to fixing related software issues.

Software Development QA Testing Functional Testing Defect identification Docker Nginx Linux