Engineering Lead - Agents

8090 Solutions•Redwood City, CA

13h•Onsite

About The Position

8090 is building a Software Factory that orchestrates the end-to-end software development lifecycle and delivers fully managed, hosted software tailored to each customer's needs. The company describes itself as a lean team working to fundamentally change how software gets made. The Engineering Lead will drive the technical direction and execution of the Software Factory agent. This is a hands-on builder-leader role combining deep expertise in agents with responsibility for developing and scaling an engineering team. The role writes and ships code while partnering with product management to turn ambitious capabilities into dependable customer experiences. Work is in person five days a week at the Redwood City, California office. The role owns agent architecture and engineering delivery, leads technical planning and execution, and develops engineers through mentorship and effective team practices. It also improves production performance using traces, experiments, and customer feedback, and partners with security and infrastructure engineers on safe deployment and operation of agents. US-based compensation is $150,000–$1,000,000 annual salary, plus stock and/or stock option awards and benefits. Compensation may vary based on individualized factors. Medical, dental, vision, and 401(k) benefits are mentioned; details are provided if and when a candidate receives an offer. Applicants must document eligibility to work in the United States and provide proof of identity, and employment is conditional on satisfactory completion of a credit and/or criminal background check. The company does not provide new work visa sponsorship.

Requirements

A record of shipping production agents with sustained, high-volume usage and measurable customer impact, including the ability to explain owned decisions, encountered failures, and delivered improvements.
Deep practical expertise in LLM-based agent systems, including orchestration, context engineering, tool integration, memory, model selection, and evaluation.
Strong software engineering fundamentals and demonstrated ability to build, debug, and operate complex production systems.
Evidence of technical leadership and mentorship that improved other engineers' work.
Clear communication, sound judgment under ambiguity, and effective collaboration with product and engineering partners.
Ability to provide documentation establishing employment eligibility in the United States and satisfactory proof of identity.
Satisfactory completion of a credit and/or criminal background check.
Must not require new work visa sponsorship from the employer.

Nice To Haves

Direct experience building coding agents or browser/computer-use agents that complete complex tasks across tools and environments.
Experience deploying agents in enterprise environments with demanding security, isolation, or customer-specific deployment requirements.
Familiarity with post-training and reinforcement learning, including judgment about when they can improve agent behavior.

Responsibilities

Own agent architecture across multi-agent orchestration, agent harnesses, prompt and context design, retrieval-augmented generation (RAG), memory, tools, and execution environments.
Lead architecture reviews and make tradeoffs between capability, complexity, latency, and cost.
Partner with product management on agent capabilities and user experience, and own engineering delivery.
Translate the shared roadmap into technical plans, milestones, and dependencies; prioritize high-impact work and remove blockers.
Regularly write and ship production code, lead code reviews, and solve difficult technical problems alongside engineers.
Build agent capabilities for reasoning about code, working across repositories and development tools, and interacting with browsers and computer environments.
Develop engineers' technical judgment, help recruit exceptional talent, and establish effective planning and review practices.
Clarify ownership, resolve technical disagreements, and build a coordinated team accountable for outcomes.
Make agent behavior observable and debuggable, with reliable recovery and escalation when tasks fail.
Use production traces, experiments, and customer feedback to improve task completion, output quality, reliability, latency, and cost.
Make progress and tradeoffs visible to leadership.
Partner with security and infrastructure engineers on permissions, sandboxing, data boundaries, and controls for deploying and operating agents in production.