2 months ago
Prague, CzechiaSenior
Responsibilities
- Embed with customers to understand their workflows and redesign functions from first principles.
- Design, build, optimize, and deploy production GenAI, LLM-based, agentic AI, and RAG systems.
- Write production code across AI, backend services, and data pipelines.
- Build evaluation harnesses, define quality criteria, instrument systems, and use evaluations to guide releases.
- Deploy maintainable systems using AWS or customer-required cloud platforms, containers, CI/CD, automated testing, monitoring, and infrastructure as code.
- Lead architecture reviews, technical design documentation, standards, technical proposals, and scoping.
- Present technical solutions and business outcomes to customer engineers, executives, and senior stakeholders.
- Transfer systems to clients, feed field learnings back into reusable blueprints, mentor engineers, and share knowledge.
- Partner with the Forward Deployed Executive to drive adoption, change management, and measurable business outcomes.
Requirements
- 7+ years of experience building and running production systems.
- Strong AI/ML foundations and the ability to reason about model failure modes.
- Production experience designing and shipping LLM applications and agentic workflows rather than demos or proof-of-concepts.
- Experience with multi-step agent orchestration, graph-based workflows, tool use, state management, and recovery from partial failure.
- Experience with LLM APIs such as Anthropic, AWS Bedrock, or OpenAI and agent frameworks.
- Production experience building and optimizing RAG systems.
- Strong engineering fundamentals with proficiency in Python and/or TypeScript.
- Hands-on AWS production experience with services such as Bedrock, Bedrock AgentCore, Lambda, ECS, S3, SQS, and ECR; GCP or Azure experience is a plus.
- Experience with containers, ECS or Kubernetes, infrastructure as code, and CI/CD for AI pipelines.
- Experience owning evaluation suites for non-deterministic systems, including ground truth, measurements, and release gates.
- Experience with model and agent monitoring, drift detection, cost control, latency optimization, model tiering, and caching.
- Production experience with Claude, Claude Code, CLAUDE.md, hooks, and skills files; MCP experience is expected and authoring an MCP server is a plus.
- Ability to make and defend architectural trade-offs, produce phased delivery plans, estimate build and operating costs, and work with ambiguity.
- B2+ English and comfort collaborating across distributed, multicultural teams.
- Preferred qualifications include experience as a founder, CTO, or engineering leader returning to individual contribution; financial services, insurance, or healthcare experience; consulting or embedded customer delivery; A2A knowledge; AWS or Claude Code certifications; GitHub Actions or GitLab CI experience; Go, TypeScript, or Rust; and Apache Spark, Apache Airflow, or Kafka.
Benefits
- Remote-friendly culture.
- Internal training and full support for Claude, AWS, and other professional certifications, plus conference attendance.
- Access to the latest AI tools and premium subscriptions.
- Long-term B2B collaboration.
- Private medical insurance or a budget for medical needs.
- Paid sick leave, vacation, and public holidays.
- Equipment and technology needed for productive work.
- Career growth and active engineer development.
Tech Stack
Apache AirflowApache KafkaApache SparkAWSAzureGitHub ActionsGitLab CI/CDGoGoogle CloudGoogle Cloud PlatformKubernetesPythonRustTypeScript
Categories
Forward Deployed
About Provectus
Provectus is an AI consultancy and solutions provider that designs and implements machine learning, data analytics, MLOps, and generative AI systems for enterprises in finance, insurance, healthcare, retail, media, and manufacturing. It sells consulting engagements and implementation services to help clients adopt AI and cloud platforms in production. Founded in 2010 and headquartered in San Francisco, the company is privately held.