AI Engineer

Необхідні навички

VisiQuate is looking for a Senior AI Engineer with a strong backend engineering background and hands-on experience building production AI systems and agentic engineering workflows.

This is not a Data Science or prompt-engineering-only role.

We are looking for a senior software engineer who has grown into Applied AI and can design, build, validate, and operate reliable systems where LLMs and AI agents are core parts of the architecture.

In addition to traditional Applied AI experience such as RAG, retrieval, and LLM-powered applications, we expect this person to be comfortable working with modern coding agents and agent harnesses such as Claude Code, Codex, or similar tools.

The ideal candidate understands how to create the engineering environment around an AI agent: provide specifications, context, tools, skills, constraints, automated validation, and review mechanisms so that the agent can safely execute substantial parts of the development workflow.

Requirements

• 7+ years of commercial software engineering experience at Senior level;
• Strong backend development experience with Java, Python, Go, C#, or a similar production language;
• Strong understanding of APIs, system integrations, databases, asynchronous processing, and distributed-system fundamentals;
• Experience designing and delivering scalable, reliable, maintainable backend systems;
• Hands-on experience building LLM-powered solutions for real business use cases;
• Practical experience with RAG, retrieval systems, vector search, knowledge-driven architectures, or document-processing pipelines;
• Experience building AI agents, tool-using workflows, automations, or multi-component AI systems;
• Hands-on experience using modern coding agents such as Claude Code, Codex, Cursor agents, or comparable systems for non-trivial engineering tasks;
• Understanding of agent harnesses: how to structure context, instructions, tools, permissions, checkpoints, and feedback loops around an AI agent;
• Experience creating or configuring skills, plugins, MCP integrations, custom tools, connectors, or agent-accessible APIs;
• Ability to design multi-step agent workflows rather than relying on single prompts;
• Experience delegating substantial engineering tasks to AI agents while maintaining quality and control;
• Ability to validate AI-generated work through automated testing, static analysis, integration tests, review agents, AI-assisted QA, and other automated verification mechanisms;
• Strong understanding of LLM failure modes, hallucinations, context limitations, tool failures, and safe fallback strategies;
• Ability to test, evaluate, monitor, and critically validate LLM and agent outputs;
• Strong ownership, debugging, code-review, and problem-solving skills;
• Ability to independently turn ambiguous product requirements into working technical solutions;
• Comfortable working across different technologies and tools when required by the problem.

Agentic Engineering Expectations

We are specifically looking for an engineer who knows how to make AI agents productive inside a real software engineering environment.

You should be comfortable with tasks such as:

• Preparing a repository or service so that an AI agent can work effectively in it;
• Defining project-level and service-level instructions, conventions, rules, and reusable skills;
• Providing agents with the right business and technical context;
• Exposing internal tools, APIs, documentation, Jira, Confluence, databases, or other systems through MCP or similar integrations;
• Designing master-agent/sub-agent or multi-agent workflows;
• Deciding what context an agent needs and how that context should be maintained or refreshed;
• Creating reusable agent tools and workflows for development, testing, debugging, documentation, and code review;
• Using automated tests and validation gates as the primary feedback mechanism for agent-generated work;
• Using review agents, QA agents, or other independent validation steps before accepting output;
• Defining boundaries for what an agent may execute autonomously and where human approval is required;
• Diagnosing why an agent failed and improving the harness, context, tools, or validation process rather than only rewriting the prompt.

We do not expect engineers to manually inspect every line generated by an AI agent.

We expect them to design systems where correctness can increasingly be verified through tests, checks, automated review, observability, and controlled feedback loops.

Буде плюсом

• Deep experience with MCP or similar agent-to-tool integration protocols;
• Experience developing MCP servers or clients;
• Experience building custom skills, plugins, tools, or connectors for AI agents;
• Experience designing coding-agent environments for large or multi-service codebases;
• Experience with multi-agent orchestration and agent-to-agent delegation;
• Experience with context engineering for long-running or complex agent workflows;
• Experience with agent memory, state management, and context compression;
• Experience with AI-assisted software development at team or organization level;
• Experience building evaluation or regression frameworks for AI agents;
• Experience with RAG evaluation, retrieval metrics, reranking, hybrid search, and grounding;
• Experience deploying AI systems or agents into production;
• Experience with cloud infrastructure, containers, CI/CD, monitoring, tracing, and observability;
• Experience working with sensitive data and designing secure, auditable systems;
• Experience in healthcare, fintech, analytics, or another data-intensive domain;
• Open-source contributions related to AI agents, MCP, developer tooling, LLM infrastructure, or backend engineering.

Пропонуємо

• Work with a brilliant, highly skilled product and engineering team that values strong technical judgment, ownership, and continuous learning;
• Access to modern AI engineering tools and infrastructure, including advanced LLMs, coding agents, agentic workflows, MCP, custom tools, and automation frameworks;
• Meaningful technical challenges at the intersection of backend engineering, AI, data, and healthcare;
• A fast-growing product company with real production use cases and room to influence technical direction;
• The opportunity to experiment with and introduce new AI tools, frameworks, and engineering practices where they bring real value;
• Flexible work schedule;
• Comfortable office in the city center, hybrid format, or remote work;
• Free English classes;
• Work equipment.

Обов’язки

• Design and develop AI-powered backend systems with LLMs as a core component;
• Build and evolve RAG pipelines, knowledge-driven applications, and retrieval architectures;
• Develop AI agents, automations, tool integrations, and agentic workflows;
• Design and maintain agent harnesses that provide agents with the right context, tools, constraints, and validation mechanisms;
• Integrate AI capabilities with existing product services, APIs, databases, and operational workflows;
• Build MCP integrations, custom tools, connectors, and reusable capabilities for AI agents;
• Create multi-step workflows where agents can research, implement, test, review, and improve their own work within controlled boundaries;
• Use coding agents to accelerate implementation of backend services, integrations, tests, migrations, documentation, and other engineering work;
• Establish automated feedback loops around AI-generated code using unit tests, integration tests, static analysis, code-review agents, and AI-assisted QA;
• Design modular, scalable, reliable, secure, and observable architectures for AI-powered applications;
• Rapidly prototype AI solutions, validate their value and reliability, and turn successful prototypes into production systems;
• Build reusable frameworks, services, skills, tools, and internal infrastructure that accelerate AI development across the engineering organization;
• Establish practical approaches for evaluation, regression testing, monitoring, and safe operation of LLM-based features and AI agents;
• Define appropriate human-in-the-loop checkpoints for high-risk or ambiguous actions;
• Collaborate with product and engineering teams to deliver reliable customer-facing functionality;
• Participate in architectural decisions, code reviews, and technical planning;
• Continuously improve how the engineering team uses AI agents for software development.

What we are not looking for

This role is probably not a fit if your AI experience is limited to:

• Using ChatGPT or Copilot for occasional code suggestions;
• Writing prompts without building the surrounding system;
• Building isolated demos without production integration;
• Using an AI agent without configuring its tools, context, validation, or permissions;
• Relying primarily on manual review to verify everything an agent produces;
• Working only with traditional ML/Data Science without hands-on LLM and agent engineering.

Про проєкт

VisiQuate is a product company building AI-powered analytics and workflow automation solutions for the healthcare domain.

Our platform integrates large volumes of fragmented healthcare data, transforms it into actionable insights, and enables intelligent workflows that improve financial performance and operational efficiency.

This role focuses on practical AI engineering and agentic software development: building reliable production systems for customers while also developing the tools, harnesses, and workflows that allow engineers and AI agents to work effectively together.

Відгукнутись на вакансію