Research Engineer (AI Systems & Evaluation)

About Us
We're a boutique engineering consultancy and AI data lab helping leading AI model providers solve complex post-training challenges. We turn ambitious research problems into practical engineering solutions that push the boundaries of AI.

That's where you come in. We're looking for a Research Engineer to build the data pipelines, evaluation environments, and engineering systems that help frontier AI models improve.

What You'll Work On
One week, you might build a sandboxed Linux environment to evaluate an AI agent resolving Git conflicts. The next, you could develop a synthetic data pipeline that tests multi-step reasoning.
Your work will include:

• AI Evaluation Systems: Build execution environments, Docker-based testbeds, and evaluation frameworks for AI agents.
• Synthetic Data Pipelines: Develop systems to generate, filter, and validate high-quality data for AI post-training.
• Internal Engineering Tools: Build annotation tools and workflows that help domain experts review complex data.
• Research-to-Engineering: Turn open-ended research challenges into reliable, executable software solutions.

What Makes You a Great Fit
We care more about how you approach difficult engineering problems than which programming language or framework you prefer.

• Engineering Fundamentals: Experience building reliable production systems, with strong debugging, testing, and validation skills.
• Systems Thinking: Understanding of APIs, asynchronous processing, concurrency, Linux, and Docker.
• Technical Curiosity: Ability to navigate unfamiliar codebases, investigate complex systems, and learn quickly.
• Ownership: Confidence in tackling ambiguous challenges and driving solutions from idea to delivery.

No research publications or model-training experience required. Strong engineering skills, curiosity, and the ability to turn complex ideas into working systems matter most.

Bonus Points

• Hands-on experience with LLM APIs, orchestration pipelines, or frameworks like LangChain and DSPy.
• Familiarity with AI post-training techniques such as SFT, RLHF, DPO, or synthetic data generation.
• Experience with AI evaluation benchmarks such as SWE-bench or OSWorld.
• Open-source contributions, competitive programming, or personal projects demonstrating strong problem-solving skills.

What You'll Get

• Work at the Frontier: Tackle real AI engineering challenges, with exposure to teams at leading AI labs.
• Grow Fast, Build More: Take ownership of challenging projects and learn alongside ambitious engineers.
• Remote Work: Work remotely as part of a team solving advanced AI engineering problems.
• Relocation Support: Optional relocation support to Yerevan, Armenia, including visa assistance and accommodation.
• Compensation & Equity: A base salary plus early-stage equity and the opportunity to share in our growth.

Built for Builders
We value curiosity over credentials, ownership over task completion, and real solutions over buzzwords.
If that sounds like you, come build what's next in AI with us!