Founding Agentic Engineer

Praxy; San Francisco, Northern (CA, KY); Hybrid; USD 150,000 - 300,000; Full time

Benefits offered by this job

Visa sponsorship

Job summary

Praxy is seeking a founding-level engineer to own the agent layer end-to-end, collaborating with the CTO to define how agentic systems are built, measured, and scaled with production-grade impact.

You will design context engineering, self-learning workflows, evals, and the agent harness, while growing the engineering team and establishing patterns for every agent built at the company.

Qualifications

• 2-9 years of software engineering experience and hands-on agentic software shipped recently.
• Familiarity with harness-centric agent development and eval-driven iteration.
• Strong knowledge of agent tracing and evaluation frameworks.
• Production-ready mindset: ship systems, not just prototypes.
• Proficiency in TypeScript; Experience with Effect-TS and PostgreSQL is a plus.

Responsibilities

• Design and own the agent layer end-to-end: context engineering, self-learning, evals, and harnesses.
• Build a knowledge base to improve future integrations.
• Develop self-healing workflows and autonomous recovery for agents.
• Set engineering patterns and standards for all agents.
• Grow the agentic engineering team and lead those who own these systems.
• Operate as a peer to the founding team and experiment relentlessly.

Skills

Software engineering

Agentic systems

TypeScript

Evaluation frameworks

Production deployment

Tools

Effect-TS

PostgreSQL

LLM tooling

Job description

This is a founding-level engineering role at a gig economy / flexible staffing platform, sitting at the intersection of production AI systems and real-world labor marketplace operations. You'll own the entire agent layer end-to-end, working directly with the CTO to define how agentic systems get built, measured, and scaled - work that has genuine, production-grade consequences.

WHAT YOU'LL DO

• Design and own the agent layer holistically: context engineering, self-learning systems, evals, the agent harness, and the meta-harness used to build and evolve new agents.
• Build a compounding knowledge base of undocumented systems so that each new integration is better than the last.
• Develop self-healing workflows - agents that detect failures in production, diagnose root causes, and recover autonomously.
• Set the engineering patterns and standards by which every agent at the company gets built, measured, and improved.
• Help grow the agentic engineering team, with a path toward leading the people who own these systems.
• Operate as a peer to the founding team and experiment relentlessly.

WHAT WE'RE LOOKING FOR

• 2-9 years of software engineering experience, with hands-on agentic software shipped within the last year.
• Deep familiarity with harness-centric agent development: context engineering and eval-driven iteration are core to how you work.
• Strong understanding of agent tracing and evaluation frameworks.
• Implementation-focused mindset - you ship working systems in production, not research prototypes.
• Proficiency in TypeScript; experience with Effect-TS, PostgreSQL, and LLM/agent tooling is a strong plus.
• Background from high-caliber engineering organizations (e.g. large tech, professional services, or developer tooling companies).

COMPENSATION & BENEFITS

Salary range: $150,000 - $300,000 USD annually. Visa sponsorship is available.

LOCATION

On-site in San Francisco, CA, United States.

Pay

COMPENSATION & BENEFITS Salary range: $150,000 - $300,000 USD