Software Engineering SMTS - Brain Team (REMOTE)

Salesforce•San Francisco, CA

17h•Remote

About The Position

The Brain team owns the intelligence layer behind Piper, Qualified’s flagship AI SDR product for qualifying leads and booking meetings. The team builds, maintains, and upgrades the agents that power Piper, including system prompts, model configuration, context, evaluations, and the harness behind the product. This role works closely with product teams to build features that let customers bring together their go-to-market strategy, CRM data, and guardrails to deliver personalized, on-brand, multimodal AI SDR experiences across websites and email. It also partners with AI platform teams to build and maintain LLM evaluations that measure Piper’s health and response quality, and assess whether changes contribute to business outcomes such as lead conversion, booked meetings, and pipeline growth. Engineers at Qualified have end-to-end ownership, from scoping and implementation through testing, deployment, monitoring, and iteration based on customer feedback. The environment emphasizes high-impact problem solving, sound engineering craft, performance, reliability, user experience, collaboration, and decisive execution. The role offers the opportunity to shape agent capabilities and tooling, own significant projects, influence architecture, and help set performance and reliability standards as the platform scales.

Requirements

At least 5 years of experience in product engineering or a related field.
Strong full-stack fundamentals and proficiency in frontend and backend development.
Customer-centric thinking and the ability to connect customer needs to the product experiences being built.
Comfort working backward from customer preferences and capturing human judgment as success criteria for an agent.
Ability to work independently, engage with product and domain experts, identify gaps, and own ideas through to production.
Understanding of agent fundamentals, including traces, context and tokens, tools, system prompts, models, latency and reasoning tradeoffs, and MCP.
Ability to troubleshoot LLM responses, identify issues in traces, and understand the strengths and weaknesses of different frontier models.
Understanding of what agents are and are not suited to do.

Nice To Haves

Experience shipping real-time and text-based LLM products.
Familiarity with Rails and React.
Hands-on work with an LLM observability system or experience building or contributing to an evaluation system.
Experience developing with agents and using coding agents effectively in daily development, including recognizing and improving low-quality agent-generated output and automating repetitive tasks.
Experience building software for marketing, sales, or go-to-market teams, particularly integrations with Salesforce and other CRM platforms.
Understanding of customer data, lead qualification, routing, and sales workflows and how to translate them into reliable product experiences.
Startup experience and rapid shipping.

Responsibilities

Build products and features that combine customers’ go-to-market strategy, CRM data, and guardrails into personalized, on-brand, multimodal AI SDR experiences for websites and email.
Collaborate with AI platform teams to build and maintain LLM evaluations measuring Piper’s overall health and response quality.
Assess whether agent and model changes contribute to business outcomes such as converting leads, booking meetings, and driving pipeline.
Tune agents based on customer preferences and translate human judgment into agent success criteria.
Own engineering work end to end, including scoping, implementation, testing, deployment, monitoring, and iterative improvement based on customer feedback.
Build, maintain, and upgrade the agents and intelligence layer that power Piper, including system prompts, model configuration, context, evaluations, and supporting harnesses.
Contribute to platform initiatives involving retrieval, model providers, and agent orchestration, and help establish performance and reliability standards.