ml engineer engineering intelligence

Armeta is an applied-AI company building engineering-intelligence products for construction and oil & gas.

• Own the product's AI/ML layer end to end, from prototype to production MVP;
• Build agentic LLM pipelines with orchestration, tool/function calling, structured output, long-context handling, model routing, and cost/latency control;
• Convert free-form engineer input into structured, validated machine representations and back;
• Design retrieval over a large domain knowledge base;
• Ship AI features supporting engineers with setup, result interpretation, and diagnosis and explanation of errors;
• Integrate the AI layer with the computation engine and product frontend via REST/WebSocket APIs;
• Build evaluation pipelines and quality metrics, including golden datasets, regression runs, and offline/online evaluation;
• Run hands-on pilots by deploying into customer environments, including on-prem and restricted-network setups, instrumenting deployments, and driving them to first real value;
• Join live sessions with domain engineers to capture task phrasing, system misinterpretations, and adoption-blocking failure modes;
• Debug in front of customers, diagnose issues live, and patch and redeploy the same day where possible;
• Turn field observations into product changes, new evaluations, golden cases, and signals about future priorities;
• Own the customer's technical relationship alongside the business analyst, including demos, integration questions, scoping, and transparent communication about current system limitations.

• 2+ Years of commercial Python experience with deep command of asyncio, typing, Pydantic, and clean architecture;
• Experience with FastAPI and production API design, including REST, WebSocket, background processing, and task queues such as Celery, RQ, or arq;
• Strong proficiency with Claude Code and its full toolset, including subagents, MCP servers, hooks, custom slash commands, CLAUDE.md, context and memory management, plan mode, skills, and headless mode;
• Hands-on production LLM experience with agent frameworks such as LangGraph, function/tool calling, structured outputs, RAG, embeddings, vector databases, fine-tuning, metrics, and evaluations;
• Experience with PostgreSQL and data handling;
• Experience taking a product to MVP in a small team, including backend, APIs, and integrations;
• Ability to hold technical conversations with practicing engineers who are not AI specialists, accept criticism without defensiveness, and remain calm when issues occur during demos;
• Comfortable debugging unfamiliar environments, working with incomplete information, and making decisions without waiting for consensus;
• Working proficiency in Russian and English;
• Willingness to travel to customer sites for pilots and deployments;
• Nice to have: technical background in engineering or applied math, numerical methods, computational geometry, scientific Python including numpy and scipy, experience with complex domain-specific data formats and engineering software, forward-deployed or solutions engineering experience, launching products from scratch with direct customer contact, and deploying into on-prem or restricted-network enterprise environments.

• Astana, Kazakhstan;
• Willingness to travel to customer sites for pilots and deployments.