Member of Technical Staff, Forward Deployed

What You'll Do

• Run technical evaluations on the customer's own workload: understand the traffic, set up the benchmark and prove the result.
• Deploy and tune each customer's model, hardware and serving configuration for their latency and throughput targets.
• Own production for your accounts. When latency moves or error rates climb, you find it and fix it, and you are who the customer hears from.
• Build the tools, integrations and runbooks that make each new deployment faster than the last.
• Work with sales on scoping and with engineering on what to build next, based on what you see in production.

What We Look for

• You are a strong engineer who has shipped and operated production systems. You can read a trace, open a profiler and find the problem yourself.
• Experience with LLM inference or serving, such as vLLM, SGLang or TensorRT-LLM, is a big plus.
• You have owned a technical customer relationship and can explain results to a skeptical engineering team.
• You are credible in a room full of engineers. You can defend a benchmark methodology and concede the point when they are right.
• You work without a spec. The problem arrives half-defined and you come back with a scoped answer rather than a list of questions.

We score every candidate on seven values

• Infinitely Resourceful
• Exceptionalism
• Unreasonable Standards
• Company Over Self
• High EQ
• Learns Quickly
• First Principles Thinker

Benefits

• Fully covered medical, dental, and vision insurance.
• Daily lunch and dinner, unlimited PTO, and parental leave.
• $1K/month housing stipend (post-tax) if you live within walking distance (0.5 miles) from the office.
• Relocation allowance and benefits.
• Covered Uber/Waymo from/to office.
• Visa sponsorship available.

How We Work

• On-site in San Francisco, five days a week.
• Small team with massive surface area and ownership.
• You operate with complete autonomy of how to solve problems.

Your success is measured by customers who move their production traffic to Wafer and stay because it keeps getting faster.