Forward Deployed Engineer

Quick Facts

• Role: Hands-on engineer working at the intersection of customer reality and product
• Focus: LLM agents, RAG, orchestration, production code, evaluations, monitoring, and adoption

Description

You will work where the product meets a customer’s real environment, starting by understanding how the organization operates and deciding what must be built. You will take ownership from early discovery conversations through a working production solution, integrating with customer data/systems while defining success metrics and driving adoption. Each implementation also serves as product discovery, feeding gaps and needs back into the product roadmap.

Responsibilities

• Run technical workshops early in projects; validate feasibility and translate needs into architecture
• Define stage-by-stage success metrics and value delivered
• Build solutions end-to-end: agents, RAG systems, orchestration, and production Python/SQL code
• Integrate with customer APIs, data silos, and legacy systems within security constraints
• Organize fragmented data in imperfect environments without a dedicated data team
• Create evals, baselines, quality gates, and monitoring that remain operational after handover
• Design reliable systems with retries, model fallbacks, cost limits, human-in-the-loop, prompt-injection resistance, and strict data boundaries
• Drive adoption and measure usage and visible results
• Turn repeated implementation patterns into reusable modules and accelerators
• Feed product gaps and customer needs back to Product and Ops
• Support sales with demos, first use-case design, and success measurement definition
• Use AI coding tools (e.g., Claude Code, Cursor) while ensuring production code quality

Requirements

• 4–6 years of software engineering with LLM-based solutions shipped to production
• Very strong Python and SQL; experience with production APIs and system integrations
• Strong knowledge of agent architectures (tool calling, context engineering, sub-agents, state management, orchestration), including MCP servers/custom tool-calling interfaces
• Ability to select an agent architecture based on task and failure modes
• Production retrieval experience: vector databases, embeddings, hybrid search, re-ranking, chunking, context management
• Data modeling and ETL/ELT pipeline experience; performance tuning; unstructured data
• Evaluation systems experience for generative solutions; tracing/observability using Langfuse or LangSmith
• Cloud deployment experience, especially Azure; containerization and CI/CD
• Application security understanding for LLM apps (OWASP, secrets management, input validation, permissions, personal data, LLM-specific attacks)
• Lead customer implementations from discovery to production; communicate technical trade-offs to non-technical leaders
• Ability to work in immature environments (spreadsheets, limited documentation, no dedicated data team)
• Strong judgment balancing scope, quality, and speed
• Ability to define implementation success metrics and demonstrate business value using customer numbers
• Daily use of agentic engineering tools and ability to review and productionize model-generated code
• Very good communication skills in Polish and English

Benefits

• End-to-end ownership of customer implementations from discovery and architecture through production and adoption
• Direct work with customers: management, technical teams, and real production environments
• Hands-on role combining engineering, product discovery, and business problem solving
• Influence on the product roadmap through implementation insights
• Opportunity to turn customer-specific solutions into reusable modules and accelerators
• AI-native, fast-moving environment focused on ownership, measurable outcomes, and continuous exploration