Staff ML Engineer, RAG & Agents

Technovate AI; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

Technovate AI in the United States is seeking a senior architect to own the technical direction of retrieval and agent work across client engagements. You will lead architecture from discovery through production, and set benchmarks before code is written.

You will partner with client engineering leadership, raise the reliability standard, and contribute reference implementations to our internal platform. This role expects deep ML/LLM experience and strong written communication.

Qualifications

• 8+ years engineering experience with 3+ shipping ML or LLM systems to production
• Depth in retrieval: hybrid search, reranking, and honest evaluation
• Experience owning a system through a compliance or security review
• Strong written communication — we write designs down before we build them

Responsibilities

• Lead architecture on retrieval and agent engagements from discovery through production
• Set the evaluation bar — design benchmarks before anyone writes retrieval code
• Review designs across the engineering org and raise the floor on reliability practice
• Work directly with client engineering leadership as the senior technical voice
• Contribute reference implementations back into our internal platform

Job description

Own the technical direction of our retrieval and agent work across client engagements. You will be the person clients upscale to when a system needs to be right, not just impressive.

• Lead architecture on retrieval and agent engagements from discovery through production
• Set the evaluation bar — design benchmarks before anyone writes retrieval code
• Review designs across the engineering org and raise the floor on reliability practice
• Work directly with client engineering leadership as the senior technical voice
• Contribute reference implementations back into our internal platform

What we're looking for

• 8+ years engineering experience, with 3+ shipping ML or LLM systems to production
• Demonstrated depth in retrieval: hybrid search, reranking, and honest evaluation
• Experience owning a system through a compliance or security review
• Strong written communication — we write designs down before we build them

Nice to have

• Consulting or client-facing delivery experience
• Published writing or conference talks on production ML
• Background in a regulated domain (healthcare, financial services)

What the first three months look like.

We write this down for every role so expectations are explicit on both sides from day one.

Pair on an active engagement, read our reference architectures, and ship a meaningful change to an existing eval harness.

Take technical ownership of one client engagement end to end, with a principal engineer available but not embedded.

Lead a discovery for a new engagement and set the architecture direction. Begin reviewing designs across other squads.

Our hiring process

• 1 Intro call — 30 minutes with a hiring lead
• 2 Technical screen — 60-minute paired session
• 3 System design — 90 minutes with two senior engineers
• 4 Team conversations — three 30-minute chats
• 5 Offer — written, within 48 hours

Fifteen days end to end. You will know where you stand at every step.

Questions first?

Email us before you apply if anything is unclear. A hiring lead will answer — not a recruiting inbox.