Agent Harness Engineer

getviktor.com

. Build the agent runtime loop so it plans, acts, and verifies its own work .

Posted 10/5/2026full-timeWarsaw • PolandMid-LevelSeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in building and modifying AI agent harnesses, with a strong understanding of system prompts, skills, and tools interactions. Capable of making technical trade-offs and designing for scale while shipping product features rapidly.

Highest-signal resume keywords

AI Agent Harness DevelopmentPython ProgrammingSystems ThinkingAgentic CodingExperience with Evals

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

AI EngineeringCustom Skills DevelopmentCLI DevelopmentMCP Server DevelopmentStress Testing ModelsTool Fluency ImprovementContext Interplay TuningTechnical Trade-Off AnalysisAutonomous Work ExecutionProduct Feature Shipping

Soft Skills

Decision MakingUser CommunicationCollaboration with Founders

Tools & Technologies

SlackTeams

Industry Keywords

Open Source ContributionsAI ToolingEval InfrastructureAgent FrameworksOrchestration Libraries

Tech Stack

Tools & technologies

Open SourcePython

About the role

Key responsibilities & impact

Build the agent runtime loop so it plans, acts, and verifies its own work
Engineer context interactions among the system prompt, skills, memory, and tools
Improve tool fluency across integrations, recall, and skill files
Expand agent capabilities including trusted memory, self-written skills, and autonomous work execution
Build evals, catch regressions, and convert failures into fixes
Ship product features to users through Slack and Teams within hours
Build whatever is needed across the company’s broad product surface
Work directly with both founders and make decisions quickly
Ship during the first week and talk to users on the first day
Build and iterate on a custom agent harness without an agent framework or orchestration library

Requirements

What you’ll need

Personally built or significantly modified an AI agent harness and can describe the trade-offs made
Understanding of how system prompts, skills, and tools interact on diverse and complex problems
Experience tuning agent context interplay
Built custom skills, commands, CLIs, or MCP servers for agentic coding
Agentic coding as a daily workflow
Experience stress-testing models and harnesses across the frontier
Systems thinking and ability to make hard technical trade-offs and design for scale
Genuine interest in how AI works, including reading research papers and having opinions about evals
Ability to work in Munich or Warsaw, or remotely in Europe
Python experience
Previous AI engineering experience, agents, harnesses, or eval infrastructure is an advantage
Previous founder or early-stage builder experience is an advantage
Open source contributions to AI tooling are an advantage

Benefits

Comp & perks

Top-of-the-market salary
Ownership/equity implied by “the kind of ownership that only exists at this stage”
Remote work in Europe
Team brought together in person a few times a year
Direct access to both founders
High-trust, low-process work environment
Opportunity to ship product features to users through Slack and Teams
Opportunity to work on frontier AI and AGI-harness technology