LLM Engineer

As an LLM Engineer at KrissDevHub, you will specialize in model optimization. Your focus will be on prompt optimization, semantic caching, routing structures, and fine-tuning models to reduce latency and API costs. You will evaluate outputs, manage token window allocations, and design fallback configurations.

What you'll do

• Implement semantic caching layers to prevent redundant model calls.
• Design routing systems that select between fast local models and frontier APIs.
• Optimize complex prompts to maximize output quality and reduce input tokens.
• Assess model hallucinations using automated test datasets.

What we're looking for

• 3+ years experience programming in Python, Go, or TypeScript.
• Deep understanding of tokenizer logic, context limits, and temperature controls.
• Familiarity with PostgreSQL, Redis, and vector caching tools.
• Outstanding analytical and benchmarking skills.

Benefits & Perks

• Remote-first flexible model.
• Competitive compensation plan.
• Premium access to all frontier LLM APIs.
• Home office budget.