# Member of Technical Staff, Forward Deployed AI Engineer at Inception Labs

- Company: Inception Labs
- What the company does: We are leveraging diffusion technology to develop a new generation of LLMs. Our dLLMs are much faster and more efficient than traditional autoregressive LLMs. Backed by AI Grant and Amplify.
- Company website: https://www.inceptionlabs.ai/
- Type: Startups (AI role)
- Level: Senior
- Location: San Mateo, United States
- Work setup: On-site
- Pay: $175K to $275K base salary per year (USD)
- Posted: 2026-05-22
- Apply by: 2026-10-12
- Apply: https://jobs.gem.com/inception/am9icG9zdDpN10tXM3_1fH8n1od_h5MV
- Page: https://www.1752.vc/careers/jobs/inception-labs-member-of-technical-staff-forward-deployed-ai-engineer/

## About the role

This role sits at the intersection of product engineering, customer implementation, evals, data collection, model optimization, and enterprise deployment ownership. You will work directly with enterprise customers to identify high-value AI workflows, collect and structure customer data, build LLM-as-judge evaluation systems, tune model and product behavior for customer-specific goals, and turn fast proof-of-concepts into production deployments.

## What they're looking for

- Enterprise customer deployments: Work directly with strategic enterprise customers to identify high-value AI workflows and turn them into production deployments
- Rapid prototyping: Build and run fast proof-of-concepts, iterating on customer requirements and technical constraints on 2-week cycles
- Production AI applications: Build full-stack AI applications, agentic workflows, integrations, internal tools, and customer-facing systems that bring Inception models into real enterprise environments
- Data collection & feedback loops: Collect, structure, and operationalize customer data to improve model and product performance on customer use cases
- Measurement and Evaluation: Define success metrics for customer deployments and design LLM-as-judge workflows, evaluation harnesses, and feedback loops for customer-specific use cases
- Model and product optimization: Tune and customize Mercury models, prompts, workflows, and system architecture to meet customer-specific performance goals