About GetReal

As the world’s leading authority on the authentication and verification of digital media, we exist to protect enterprises, governments, and media from AI-powered deception, deepfake fraud, and identity manipulation — because trust in digital content is no longer guaranteed.

We bring together a unique blend of multi-disciplinary domain expertise to protect organizations from a new generation of attack vectors and rebuild trust in digital communications. We move fast, debate hard, and hold each other to a high bar — and we’re looking for a Senior Agent Engineer who is sanely principled about AI, owns outcomes end-to-end, and treats agentic systems as an engineering problem worth solving properly.

The Role

We are hiring a Senior Agent Engineer to own both the implementation and strategic direction of agentic systems at GetReal — inside the company and inside our products. You will ship the agents that change how we build, support, and sell, and help shape the agentic capabilities our customers experience.

You will also build the subsystems that don’t exist yet — the tool layers, eval harnesses, memory and retrieval primitives, guardrails, and observability agents need to work in production at GetReal’s scale and risk profile.

We care deeply about being sanely principled about AI: energized by what agents can do, clear-eyed about where they can fail, what they cost, and what they should not be trusted with. Allergic to hype, and just as allergic to dismissiveness.

This role is in-office in Atlanta, GA or Austin, TX.

What You’ll Do

• Design and ship agentic systems — single- and multi-agent, tool-using, evaluable — for internal teams and for customer-facing product surfaces.

• Stand up the supporting subsystems agents need in production: tool/MCP layers, memory, retrieval, web crawling and data acquisition, guardrails, evals, tracing, and cost/latency observability.

• Build internal automations that materially change how GetReal operates across engineering, research, GTM, support, and ops — and measure their actual impact.

• Own GetReal’s point of view on where agentic AI belongs in the roadmap — internal vs. product, build vs. buy, model and framework choice, sequencing.

• Run rapid, honest experiments to validate or kill agentic ideas before they consume serious engineering investment.

• Operate within real constraints — security, compliance, customer data sensitivity — and design accordingly rather than around them.

What We’re Looking For

• Sanely principled about AI. Strong views grounded in things you have actually built and measured — not vibes, and not reflexive skepticism.

• Owns outcomes end-to-end. Takes ambiguous problems, decomposes them, ships, instruments, and stays with the work until it actually works.

• Builds what isn’t there yet. Comfortable being the first person to stand up a subsystem inside real engineering constraints.

• Strong fundamentals. Production code, reliability, latency, and cost as first-class concerns.

• Hands-on with modern agent stacks. Tool use, MCP-style calling, planning, retrieval, structured outputs, multi-agent orchestration, and evals — not just demos.

• Culture fit. Low-ego, opinionated, bias to action, comfortable being challenged and challenging back. Wants to be on a small, senior, in-office team that ships.

Technical Competencies

• Proficient in Python; comfortable in at least one of TypeScript / Go.

• Production experience with frontier LLM providers (Anthropic, Google, and others) and the trade-offs between them.

• Deep familiarity with single- and multi-agent orchestration patterns — tool use, MCP, planner-executor loops, sub-agent delegation, structured outputs, retrieval augmentation.

• Experience designing and running evaluations for both deterministic and non-deterministic systems — RAGAs, F1 / precision / recall, golden traces, offline eval sets, online metrics, and regression tracking.

• Experience building web crawling and data acquisition pipelines — bespoke crawlers, headless browsers, and third-party APIs — for use cases where the data isn’t handed to you (verification, enrichment, evidence gathering).

• Comfortable with production AI infrastructure (serving, tracing, observability, rollout) and cloud infra (AWS preferred).

• Bonus: multimodal (audio / video / image) experience, business-system integrations, or guardrail / red-team work.

Why GetReal

Our mission shapes what we build and how we build it. Here’s what that means for this role:

• You’ll own the agentic strategy from the ground up — no inherited mess, no politics around what you can change.

• The problems are genuinely hard: building production agents inside real security, compliance, and accuracy constraints, where being wrong actually matters.

• A small, senior team means you’ll work directly alongside the engineers, researchers, and product leaders shaping where agentic AI lives in our product.

• We use what we build every day — what you ship internally is a candidate to graduate into customer-facing capabilities.

• Your work will be visible: the leverage of agentic AI inside a company our size is felt across the team — and the confidence we get from being principled about it matters when the product is protecting people from fraud.

We celebrate diversity and are committed to creating an inclusive environment for all employees. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or