
ML Engineer, AI/LLM
Posted 12 hours ago

Posted 12 hours ago
This is a fully remote position, open to applicants in Uzbekistan.
• Take charge of AI design in client projects, including chat flows, knowledge searches, safety checks, model selection, and testing.
• Develop retrieval (RAG) pipelines utilizing embeddings, hybrid search, and reranking on PostgreSQL with pgvector.
• Create prompts, context, and multi-step chat flows that incorporate agents, tool usage, and streaming replies.
• Collaborate with open-weight models like Qwen, Llama, and Gemma, deployed through LiteLLM or another OpenAI-compatible gateway and Azure AI Foundry.
• Establish evaluation metrics: create test sets, implement automated quality checks, and conduct side-by-side comparisons of models based on quality, speed, and cost.
• Integrate safety measures from the outset: develop guardrails, safety classifiers, content moderation, and defenses against prompt injection.
• Select the appropriate model based on test outcomes and clearly articulate the rationale to non-technical clients.
• Review AI code produced by other developers and set standards for LLM development practices.
• Participate in client technical calls to address their AI inquiries directly.
• Minimum of 4 years in software engineering or machine learning, with at least 2 years dedicated to developing LLM applications used by real users; prototypes and demos alone are not sufficient.
• Proficient in Python.
• Practical experience with RAG in production, including embeddings, hybrid search, reranking, and vector databases.
• Expertise in prompt and context engineering, as well as multi-step chat flows (orchestration, agents) with streaming responses.
• Demonstrated experience in real evaluation work: you've created test sets and automated quality checks, using them to compare models on quality, speed, and cost.
• Background in AI safety: including guardrails, safety classifiers, content moderation, and protection against prompt injection.
• Comfortable working within a TypeScript / Node.js codebase in collaboration with the backend team.
• Ability to explain model choices and their trade-offs to non-technical clients in straightforward terms.
• Proficient in spoken and written English; you will produce design documents and participate in client technical discussions.
• Regularly utilize AI tools in your engineering work as part of an AI-augmented workflow.
• Arabic speaker or experience in developing LLM products in Arabic or other non-English languages is a significant advantage.
• Familiarity with NestJS.
• Experience in fine-tuning with LoRA or QLoRA, and self-hosting models on GPUs (e.g., vLLM).
• Knowledge of speech-to-text, text-to-speech, or voice AI technologies.
• Experience with healthcare, mental health, or other regulated data, including compliance with data residency regulations.
• Familiarity with Azure, particularly Azure AI Foundry.
• Full-time, in-house position on the Bolder Apps team.
• Lead AI design on actual products within a regulated environment, avoiding mere prototypes.
• Consistent flow of AI projects following the initial one.
• Genuine ownership over model selection and architecture, free from a rigid framework.
• Direct access to decision-makers, with no unnecessary layers between you and those responsible for outcomes.
• A network of peers including engineers, designers, and PMs collaborating on interrelated projects.
Extend
OhioHealth
Capital One
Software Mind
Get handpicked remote jobs straight to your inbox weekly.