SLM Engineer

Kforce has a client in NYC that is seeking an SLM Engineer who will be designing, fine-tuning, and deploying small to mid-sized language models (SLMs) using AWS Bedrock. This role is centered on hands-on model customization, optimization, and production integration within an AWS-native environment. Ideal Candidate Profile:

• Background: ML Engineer, AI Engineer, or Software Engineer focused on GenAI applications
• Strengths: Fine-tuning, Bedrock implementation, and production deployment
• Languages: Python required
• Environment: AWS-first, strong familiarity with Bedrock and cloud-native AI patterns
• Mindset: Builder focused on delivering usable, scalable AI solutions

• Successfully deploys Bedrock-powered models into production environments
• Fine-tunes and optimizes models for domain-specific accuracy and performance
• Improves response quality, latency, and cost efficiency over time
• Integrates AI capabilities seamlessly into real-world applications

• Optimize models for latency, cost, and performance in real-world environments
• Run experiments and iterate based on measurable improvements
• Build and deploy APIs to serve models in production
• Integrate Bedrock-powered models into applications and enterprise systems
• Develop pipelines for training, evaluation, and deployment workflows
• Partner with application and data teams to embed AI into business processes
• Strong hands-on experience with AWS Bedrock for building and deploying generative AI solutions
• Experience working with foundation models available through Bedrock (Anthropic, Meta, etc.)
• Ability to configure, customize, and optimize models within Bedrock for specific use cases
• Familiarity with broader AWS ecosystem: SageMaker, S3, Lambda, ECS/EKS
• Experience deploying scalable, production-ready AI services in AWS
• Practical experience working with language models and transformer-based architectures
• Experience adapting models for domain-specific use cases such as summarization, Q&A, or classification
• Strong understanding of how to evaluate and improve model outputs
• Familiarity with prompt engineering and instruction tuning techniques
• Hands-on experience with fine-tuning models using Python-based frameworks (Hugging Face, PyTorch, etc.)
• Experience with parameter-efficient fine-tuning techniques: LoRA, PEFT, adapters
• Ability to prepare and manage training datasets including cleaning and labeling
• Experience with ML lifecycle management: Model versioning, monitoring, retraining workflows
• Experience with CI/CD pipelines for ML and AI workloads
• Familiarity with containerization (Docker) and deployment in ECS/EKS
• Understanding of GPU and accelerated compute environments