Solution Architect (Agentic Development Lifecycle)

About the Role
We are looking for a hands-on Solutions Architect to lead the agentic transformation of how Geniusee builds software. You will own the Solution Architecture role inside the Geniusee AI Factory, our in-house platform of contracts, gates, shared memory and Claude Code role plugins that takes an engagement from the first client call to working software, and you will extend the factory itself: agents, skills, worker packages, gates and the development pipeline. Then you will make it stick: train our engineers, BAs, QAs and PMs, and mentor the teams through the change.
This is a Center of Excellence role reporting directly to the CTO. You will split your time roughly in three: building agentic pipelines on live projects, teaching, and mentoring. You will own the roadmap and come with decisions, not questions.
This is a principal-level role for someone at the top of the agentic engineering field. You live on the frontier: you know what shipped in Claude Code, the Claude Agent SDK, MCP and the competing stacks last week, you have opinions on which agent architectures actually hold up in production and which do not, and you have the engineering depth to prove it on a brownfield codebase with a client watching. We expect you to set the bar for 300+ engineers, not to learn alongside them.

Requirements:
● 7+ years in software engineering, with 2+ years as a Solutions Architect or Tech Lead delivering commercial systems for US/EU clients. Strong background in microservices, event-driven architectures, AWS, and working with legacy/brownfield codebases.
● Strong TypeScript/Node.js (platform core), plus hands-on fluency in at least one enterprise stack (Python, Java/Kotlin, .NET, or React).
● 12+ months of daily, production-grade development using coding agents (Claude Code required; Cursor or Codex as a plus) with shipped features and at least one measurable agentic SDLC rollout across a team of 5+ engineers.
● Deep practical knowledge of multi-agent patterns (single loop, planner/executor, orchestrators, swarms, agentic RAG) and hands-on experience building custom MCP servers/clients, designing robust tool schemas, auth, and sandboxing.
● Feature-level mastery of the Anthropic stack (Claude Code, Agent SDK, headless runs, hooks, subagents, prompt caching, memory, permission modes) and working knowledge of alternative stacks (OpenAI Agents SDK, LangGraph, CrewAI) to defend model routing by cost, latency, and capability.
● Hands-on experience with spec-driven delivery (OpenSpec, Spec Kit, BMAD) and agent quality engineering: automated evals on real codecases, LLM-as-a-judge, tracing (token spend, latency), prompt regression testing, and security guardrails (injection, secret leakage).
● Proven ability to run technical discovery, build architectural vision (ADRs, NFR targets), construct WBS-based estimates, and defend technical choices directly to clients.
● Demonstrated experience running technical workshops, upskilling engineers, assessing competencies, and driving change. Fluent English (B2+) for client demos and debates.
Nice to have:
● AWS Certified Solutions Architect (Professional) or Machine Learning Specialty.
● Public open-source contributions, custom plugins, MCP servers, or technical articles around agentic engineering.
● Experience with design-to-code pipelines (e.g., Figma MCP, visual regression, Playwright).
● Experience delivering projects under outcome-based or fixed-price models.
● Exposure to compliance frameworks in LLM contexts (SOC 2, HIPAA).


Responsibilities:
● Own and evolve the Solution Architecture role, plugins, worker packages, and gate checks inside our internal agentic platform; contribute to CLI, controller, and CI workflows.
● Act as the hands-on SA on presales and discovery phases for US/EU clients (ballpark estimates, solution architecture briefs, ADRs, client-facing readiness packages).
● Lead hands-on pilot implementations across 2–3 delivery projects simultaneously, auditing codebases, setting up CLAUDE.md and MCP tooling, and proving cycle-time improvements.
● Establish company-wide standards for spec-first development, reusable agent templates, and automated review/testing pipelines that feed directly back into the agent loop.
● Build and deliver multi-level training programs (Foundation to Lead), mentor an internal cohort of tech champions, and run technical office hours.
● Define and track agentic delivery metrics for the CTO (cycle time, code delivered through agents, defect rates, and EVM impact).

What you will get:
Career growth & projects
●Competitive salary based on your skills and experience, and a benefits package.
● Regular performance appraisals to support your career development.
● Professional development support: online courses, certifications, and study compensation (including English courses).
● A modern tech stack across challenging, long-term projects.

Health & well-being
● Paid vacation (18 working days), public holidays (11 days), and sick leave (15 days).
● VIP medical insurance or sports coverage.

Work environment & flexibility
● Remote work from anywhere in the world or in one of our offices in Kyiv, Lviv, or Warsaw.
● Necessary equipment to perform your work.

Culture & community
● A collaborative and open culture where contributions are recognized.
● Regular corporate online & offline activities.
● Geniusee Charity Fund, making a difference beyond tech.