We are hiring a Senior AI Engineer to own our AI’s ability to understand, respond to, and act on user queries from the ground up. This is not a research role or a “prompt tinkering” side project. You will build agentic LLM systems that can connect to external tools and APIs, structure their reasoning, and execute actions reliably.

You’ll be hands-on with prompt engineering, LLM frameworks, and agent design, while also applying software engineering rigor to make these systems stable, scalable, and useful. The right candidate moves fast, learns new methods as the field shifts, and always drives toward business outcomes, not demos or academic novelty.

We don’t believe in overwork or equating hours with outcomes. What matters is results tied to business and customer outcomes—nothing else.

This is a hybrid role, 3–4 days per week in-office in Mission Viejo.

What You’ll Do

Design and implement LLM-powered agents that can reason, decide, and act via connected APIs, databases, and external systems.
Own the prompting, fine-tuning, and orchestration that make these systems intelligent and usable.
Apply strong engineering principles (modularity, testing, observability) so the AI’s actions are reliable in production.
Continuously scan the landscape for new frameworks, libraries, and techniques, and integrate what’s useful.
Collaborate with product and leadership to ensure progress is tied directly to business results.
Document decisions and trade-offs, ensuring clarity for current and future teammates.
Proven ability to deliver usable systems under startup constraints: full ownership of outcomes, small teams, fast pivots, and delivering usable systems with limited resources.
Work in a modified agile framework based on your direct input (no dogma or purist approach) with daily standups and project management coordinated between you, the COO, and two other engineers (frontend and backend).
Take strategic and product guidance from a cofounder who will be in-office with you in Orange County. The COO supports your growth and ensures you have what you need to succeed, while the cofounders focus on product excellence
You'll occasionally join customer calls to hear feedback and problems directly from users.

Must-Haves

Strong, validated expertise in software engineering with expertise shipping production systems from beginning to end.
Deep hands-on work with LLMs (OpenAI, Anthropic, open-source stacks, LangChain/LlamaIndex/transformers, etc.) and expertise in building agents using them.
Experience building multi-step, tool-using agents that act across varied systems, not just single-action chatbots
Proven ability to pick up new technologies quickly and apply them pragmatically.
Ability to turn messy, ambiguous problems into solvable projects, with clear scoping, regular updates, and visible progress for stakeholders
Excellent communication: clearly defines projects, can explain assumptions, trade-offs, and progress clearly.

What We Value

We hire for mental models, not buzzwords. The right candidate will embody:

Bias to Progress → Keeps work moving, persistent in ambiguity, avoids getting stuck.
Business Results First → Prioritizes impact over complexity, identity, or “cool tech.”
Simplicity Bias → Builds the leanest path to the outcome, avoids overbuilding.
Ownership → Takes responsibility, integrates feedback, doesn’t deflect blame.
Adaptability → Learns new tools fast, evolves methods as the field shifts.
Collaborative Pragmatism → Can disagree respectfully, align quickly, and execute.

Nice-to-Haves

Experience with retrieval-augmented generation (RAG) or vector DBs.
Exposure to evaluation frameworks for LLMs to ensure reliability scoring and reduce hallucinations.
Experience in startups or lean environments where speed and iteration matter.

Compensation and Growth

$150,000 + Equity

Health, Dental, and Vision Insurance + Dependents

Four Weeks Paid Time Off (We try to encourage you to take time off)

Interview Process

At Humoniq, we design hiring to be fair, predictive, and respectful of your time. You’ll have the chance to show what you can do on your own terms, choosing when and where, so you’re able to perform at your best.

The data is clear: resumes are one of the weakest predictors of job performance (Hunter & Schmidt’s meta-analysis, ~0.1–0.2 correlation). They’re noisy, biased toward pedigree, and reward formatting skills.

That’s why we combine structured interviews with research-backed assessments.

Fairness: Interviews can sometimes reward confidence over skill. Assessments give every candidate the same chance to show their abilities, leveling the playing field.
Effectiveness: Decades of studies show assessments are stronger predictors of job success than interviews alone. They help us focus on real skills, not just first impressions.
For You: This means you won’t be judged on your resume, background, accent, or personality quirks, but on what really matters is your ability to do the work.

Our entire process will take less than 8 hours, and will not require any 'take home’ work tests from you. There will be an in-person component as part of the last interview.

Humoniq (YC S25) AI Jobs

Latest AI Jobs at Humoniq (YC S25)

Senior AI Engineer