Member of Technical Staff, ML Product Engineering
AI summary of the role
Inception Labs seeks an experienced ML engineer/scientist to train and deploy diffusion-based LLMs (Mercury) in production, bridging research and real-world applications.
What you’ll do
- Design, develop, and optimize models for production use cases
- Partner with customers to translate requirements into technical solutions
- Implement post-training approaches for generative AI, including agentic workflows
- Build data preprocessing pipelines, model evaluation, and alignment for enterprise use
What you’ll bring
- BS/MS/PhD in CS, ML, or related field (or equivalent)
- 5+ years of ML project experience in PyTorch or equivalent
- Strong familiarity with transformers and core LLM concepts (autoregressive pretraining, instruction tuning, LoRA, KV caching)
- Experience training and fine-tuning LLMs
Technologies
PyTorch · transformers · LLM · LoRA · KV caching · GPU/TPU · Docker · vLLM · SGLang · TensorRT
Source and classification
Production engineering · Evidence for this classification:
Inception creates the world’s fastest, most efficient AI models. Our Mercury model is the world’s fastest reasoning LLM and first commercially available diffusion LLM, delivering 5x greater speed and efficiency than today’s LLMs, with best-in-class quality. We are the AI researchers and engineers behind such breakthrough AI technologies as diffusion models, flash attention, and DPO. The Role We seek experienced engineers and scientists to bridge the gap between research and real-world applications by training and deploying our diffusion large language models. You'll build our core product offerings, partner with customers, and ensure our models perform reliably at scale in production environments. Key Responsibilities Design, develop, and optimize our models for production use cases. Partner with customers to understand their requirements and translate them into technical
More from the job description
Inception creates the world’s fastest, most efficient AI models. Our Mercury model is the world’s fastest reasoning LLM and first commercially available diffusion LLM, delivering 5x greater speed and efficiency than today’s LLMs, with best-in-class quality. We are the AI researchers and engineers behind such breakthrough AI technologies as diffusion models, flash attention, and DPO. The Role We seek experienced engineers and scientists to bridge the gap between research and real-world applications by training and deploying our diffusion large language models. You'll build our core product offerings, partner with customers, and ensure our models perform reliably at scale in production environments. Key Responsibilities Design, develop, and optimize our models for production use cases. Partner with customers to understand their requirements and translate them into technical solutions. Implement innovative approaches for post-training generative AI models, including agentic workflows. Work on data preprocessing pipelines, model evaluation, and alignment to enterprise use cases. Contribute to the deployment and maintenance of models in production environments. Collaborate with product teams to design and implement customer-facing ML features. Qualifications BS/MS/PhD in Computer Science, Machine Learning, or a related field (or equivalent experience). At least 5 years [... source excerpt omitted ...] ility to explain technical concepts to non-technical stakeholders. Preferred Skills Expertise in data engineering and synthetic data generation for LLMs. Knowledge of MLOps and production-level deployment workflows. Experience with LLM serving frameworks like vLLM, SGLang, or TensorRT. Experience with cloud platforms (AWS, GCP, Azure). Experience with model quantization and optimization techniques. Compensation The annual base salary range for this role is $200,000 – $350,000 USD. Final compensation is determined based on experience, skills, and qualifications. Equity and benefits are included in the total package. Why Join Inception Work with World-Class Talent: Collabor
Employer postings · Data from · Sources