Applied AI Engineer

Twenty is seeking a mid-level Applied AI Engineer for an in-office position in its Arlington, VA office to help build the next generation of cyber technologies that protect democracies worldwide. We're looking for someone with 4 years of experience in developing and supporting ML and AI applications. In this role, you'll build, tune, and deploy language models that enhance our mission-critical cyber capabilities. You'll create and curate specialized datasets, fine-tune models for specific use cases, and implement efficient retrieval systems to augment AI reasoning. You'll join a world-class product and engineering team that delivers mission-critical solutions for U.S. national security, working in both cloud and on-premises environments to build AI systems that operate at machine speed. If you're passionate about applying cutting-edge AI techniques to solve complex technical challenges while making a direct impact on national security, we want to talk to you.

About The Company

At Twenty, we're taking on one of the most critical challenges of our time: ensuring democracies prevail in the digital age. We develop revolutionary technologies that operate at the intersection of cyber and electromagnetic domains, where the speed and complexity of operations exceeds human cognition. Our team doesn't just solve problems – we deliver game-changing outcomes that directly improve national security. We're pragmatic optimists who know that while our mission of defending America and its allies is challenging, we can succeed.

Role Details

• Create, clean, and maintain high-quality training and evaluation datasets
• Fine-tune language models ranging from small specialized models to medium foundation models
• Develop and optimize model serving infrastructure for production environments
• Develop retrieval-augmented generation (RAG) systems to enhance LLM capabilities with external knowledge
• Implement prompt engineering strategies to optimize model outputs for specific applications
• Integrate AI capabilities into applications using frameworks like LangChain and CrewAI
• Design and implement evaluation frameworks to measure model performance
• Collaborate with cross-functional teams to identify use cases where AI can add business value
• Create technical documentation for AI systems and processes
  
  

Qualifications

Technical Skills & Experience

• 4 years of professional software development experience with full-stack applications utilizing ML or AI
• Strong skills in Python and deep learning frameworks (PyTorch, TensorFlow, or JAX)
• Proficiency in LLM post-training techniques especially RL-based methods (continued pre-training, SFT, RLHF, DPO, PPO, GRPO)
• Proficiency in relational, graph, and vector database concepts
• Knowledge of dataset curation, cleaning, and preprocessing methodologies
• Familiarity with LLM and agent evaluation metrics and testing procedures
• Experience with AI orchestration frameworks like Pydantic-AI, LangChain/LangGraph or CrewAI
  
  

Education

• Bachelor's degree in Computer Science, Software Engineering, or related fields, or equivalent practical experience
  
  

Security Requirements

• Must be eligible to obtain and maintain a U.S. Government security clearance
  
  

Distinguishing Qualifications

• Previous experience deploying models to production environments
• Understanding of prompt engineering and model alignment techniques
• Experience with distributed training systems
• Knowledge of model quantization and optimization techniques
  
  

Additional Skills

• Machine learning operations (MLOps), LLMOps, AgentOps
• Vector databases (pgvector, ChromaDB, Pinecone, Milvus, Weaviate)
• Model serving technologies (vLLM, TensorRT, ONNX)
• Containerization and orchestration (Docker, Kubernetes)
• Experimentation tracking and version control for ML (dataset versioning, prompt versioning, model versioning)
• Cloud platforms (AWS, GCP, or Azure)