Middle AI Agent Engineer / Agentic AI Engineer

Anadea | Tensorway is an international distributed team with more than 20 years of experience creating corporate and startup solutions (web, mobile, AI). We predominantly work remotely from Ukraine and Europe, with our office located in Dnipro.
You understand the risks of overreliance on coding agents and recognize the importance of taking full ownership and responsibility for the outcomes of your work, whether the code is LLM-generated or written by you. We’re looking for a proactive, self-driven individual who takes initiative, and proven leadership experience would be a strong plus.

Required Skills
Hard skills:
Python: proficient knowledge of the language for development, debugging, and code maintenance.
LLM Frameworks: practical experience with LLM-applications and agent frameworks, building agent loops, multi-agent systems, RAG, context management, MCP.
Production Safety: guardrails and safety controls, human-in-the-loop (confidence calibration for optimal delegation).
Assisted SDLC: creating sharable coding agents configurations, building reusable modular context, employing agents at all stages of the SDLC.
API & Testing: ability to design REST/GraphQL interfaces and write unit/integration tests with Git as the foundation of the workflow.
Software Development: concurrent programming and database systems, software design principles and patterns.
Monitoring Tools: experience with LangSmith, OpenTelemetry, MLflow for tracking and analyzing agent systems.
Eval Pipelines: evaluating LLM-based workflows and non-deterministic agent trajectories, using dedicated frameworks and approaches, defining metrics.
DS and ML foundations: training, evaluation, bias, and visualization.
DevOps Basics: Docker containerization and CI setup (GitHub Actions/GitLab CI) for “green” builds and agent testing.
Cloud Platforms: familiarity with AWS, GCP, or Azure for deployment, scaling, and load balancing.

Bonus Knowledge and Preferred Qualifications:
Rust or another systems language for high-performance services.
Fine-tuning and Local Models: production experience with quantization, fine-tuning, and working with RLHF approaches.
Experience with TTS, STT, or real-time models, or with computer vision, OCR.
Certifications such as Claude Certified Architect (Professional or Foundation).
Experience with compliance requirements (GDPR, HIPAA, or SOC 2).

Why Join Us?
Conditions and Benefits
🌍 Remote-first & flexible work
Fully remote working model.
Access to an office if there is one available in your location.
Monday–Friday, 40-hour working week.
Flexible working hours: schedule options between 08:00 and 19:00, including a 1-hour lunch break (the only requirement is a minimum of 5 hours of overlap with the team)
🏖️ Vacation & sick leave

20 paid working days of vacation during the first year.
10 paid sick days per year (for childcare would also be covered), compensated at 100% of salary.
Public holidays are days off by default.
❤️ Health & family benefits

Medical insurance or medical insurance compensation.
Company gifts for:Birthday
Wedding
Birth of a child

Paid lunches at the office.
Family days etc.
📑 FOP / Individual Entrepreneur Support if needed
🔄 Paid Bench period between projects
🤝 No time-tracking software — transparent and trust-based time reporting.
🎁 Referral Program

If you meet the key requirements and want to work on scalable autonomous agent systems, please send your resume and project links!