TL;DR

AI Engineer (Telegram Mini Apps): Developing AI-first products inside Telegram, including LLM-assistants and AI-driven gaming and automation platforms with an accent on integrating LLMs, building RAG systems, and developing AI assistants. Focus on optimizing latency, inference pipelines, and integrating AI functionality into Telegram Mini Apps.

Salary: 450 000 – 550 000 ₽

Компания

Apex Play builds AI-first products inside Telegram.

Что делать

  • Integrate LLMs (OpenAI / Anthropic / open-source models) and optimize routing and cost.
  • Build RAG-systems: embeddings, semantic search, vector DB.
  • Develop AI-assistants: multi-turn dialogues, memory, tools/tool-calling, guardrails.
  • Optimize latency and inference pipeline (caching, batching, streaming).
  • Write backend on Python (FastAPI) for AI-functions (API, background tasks, queues).
  • Integrate AI-functionality in Telegram Mini Apps: WebApp auth (initData), Bot API, webhooks, Telegram WebView limitations, UX-flow inside Telegram.

Требования

  • Опыт разработки Telegram Mini Apps / Telegram Web Apps в production (not bots “in vacuum”, but Mini App + backend/AI integration inside Telegram).
  • Real experience with LLM APIs and/or open-source models.
  • Experience with RAG and vector DBs: Qdrant / Pinecone / Weaviate.
  • Python (FastAPI), confident server-side development.
  • PostgreSQL, Redis (cache/sessions/rate limiting).

Хорошо, если есть

  • Production AI-products with real users and metrics.
  • Streaming responses, voice/vision models.
  • Highload: queues (Kafka/RabbitMQ/Celery), horizontal scaling, observability.
  • Fine-tuning / LoRA, understanding inference on GPU, cost optimization.

Культура и преимущества

  • High autonomy and influence on decisions and product portfolio.
  • Fast releases and clear KPIs.
  • High income level - fix + bonuses (we focus on your expectations).
  • Fully remote work format. You can work from anywhere in the world.
  • Any type of employment: full, part-time, project.