Loman AI is transforming how restaurants handle phone calls with an AI-powered voice platform built specifically for the restaurant industry. Our technology answers calls 24/7, takes orders, books reservations, answers menu questions, and integrates directly with restaurant POS and reservation systems to create a seamless guest experience. By eliminating missed calls and hold times, Loman helps restaurants increase revenue, improve operational efficiency, and deliver better hospitality at scale. We’re building one of the most advanced voice AI products in hospitality and partnering with restaurants ranging from independent operators to large multi-unit brands. Our team combines deep expertise in AI, SaaS, and restaurant technology with a fast-moving startup culture focused on innovation, execution, and customer impact. At Loman, you’ll have the opportunity to help shape the future of how restaurants communicate with their guests while working on cutting-edge technology that solves real-world problems for one of the world’s largest industries.

About Loman

Loman is the #1 rated Voice AI for restaurants. Our AI phone agent answers 100% of calls, 24/7 — taking pickup and delivery orders, processing payments, booking reservations, and answering menu and FAQ questions, with clean tickets syncing straight into the restaurant's POS. Built in Austin. Used by some of the country's top operators.

ARR is ramping fast. We're investing heavily in the post-sale operating model — segmentation, health, cadence, and expansion — and CS is the function that turns Loman from a great product into a long-term revenue engine.

About the Role

As a Forward Deployed Engineer in the Customer Success team, you are Loman’s customer-embedded technical partner, focused on maximizing the strategic outcomes of our deployed Voice AI agents across all commercial customers. This position balances continuous, high-level post-launch support with the rapid development of innovative, bespoke tooling to scale customer enablement effectively.

You are a resourceful, gritty, and adaptable engineer who views customer success as a personal engineering problem. Instead of waiting for product roadmaps, you leverage your technical expertise to rapidly scope, build, and deploy bespoke solutions - from integrations and micro-features to tailored agent flows - to unblock mission-critical workflows for top-tier operators. This is one of the highest-leverage seats at Loman, where your direct impact solidifies the company’s reputation as 'the AI vendor that actually shows up.

Key Responsibilities

As a Forward Deployed Engineer, you will contribute your creativity and problem-solving to define how we deliver business value to the customer with increasing efficacy and efficiency. You can expect to:

Drive Continuous Optimization and Value Creation: Own the post-launch performance of each customer's agent, continuously tuning its behavior, refining prompts, and optimizing rule sets using real call data, transcripts, and observability traces to ensure measurable improvements month over month.

Engineer Bespoke Solutions for Scale: Build custom solutions when customers encounter pain points the standard product does not yet handle. This includes shipping custom automations, scripts, or agent flows—whatever it takes—to ensure customer needs are met at speed, not roadmap speed.

Own agent experience: Accuracy, naturalness, brand voice, edge-case handling. What the operator hears on the phone is your job.

Go above and beyond. Move at customer speed, not roadmap speed. When a top operator needs something to keep working, you make it work - and you don't wait for the sprint.

Support technical onboarding. POS integration, menu setup, agent configuration, factory evals, pre-launch QA. Partner with the Onboarding Specialist, who owns the customer relationship.

Feed Engineering. Patterns from your customers - bug clusters, repeat pain points, frequent custom asks - go back into Product and Engineering planning so the whole base benefits.

Document everything. What you built, what you tuned, what's escalated, what's custom. Nothing should die when you take a vacation.

What Success Looks Like

In your first 30 days you've ramped on the stack - Langfuse, Retell, the codebase, the POS integrations, factory evals - taken your first set of customer accounts, and made measurable improvements to at least one agent post-launch.

In your first 60 days you're managing your book of customers; agent quality is trending up across them; you've shipped your first bespoke build for a specific customer; and technical onboardings are routing through you.

In your first 90 days your customers describe you (in their words, not ours) as 'their engineer'; agent quality is materially better than at handoff; you've shipped multiple custom builds; NRR signals on your accounts are healthy; and Engineering has picked up at least one of your patterns into the core product.

Required Experience

• 2+ years as a software engineer (full-stack, backend, ML/AI, infrastructure, or similar) — ideally with customer-facing or solutions-engineering experience.
• Comfortable in production debugging across logs, traces, webhooks, and APIs. You can read a Langfuse-style trace, follow a webhook flow, and read code you didn't write.
• Hands-on experience with LLM / AI agent systems - prompt engineering, agent behavior, state machines, evals.
• Track record of shipping things end-to-end against a deadline that mattered to a real customer.
• Strong written and verbal communication - comfortable on a customer call and in a code review in the same hour.
• Bias for action; not waiting for permission to ship something a customer needs.
• Highly organized; able to manage multiple live customer-facing engagements without things falling through.

Nice-to-Have Experience

• Experience as a forward-deployed / solutions / customer engineer at an AI or vertical-SaaS company.
• Direct experience with voice AI infrastructure (Retell, Vapi, ElevenLabs, Twilio Voice) or agent observability tools (Langfuse, LangSmith, Helicone, Arize).
• Experience integrating with restaurant POS or ordering platforms
• Experience designing and running evals for LLM systems.

Ideal Candidate Profile

You treat your customers like an engineering project. You don't wait for a ticket queue - you go look at the call logs, find the agent's weak spots, and fix them before the customer raises them. You'll get on a call with an operator, figure out what they actually need, build it that week, and ship it.

You read traces and logs the way other engineers read code. You write clean repros, you escalate with specificity, you don't escalate before you've actually looked. You write down what you build so the team can scale it. You don't believe a problem is solved until the customer is back to revenue.

You want to be at a company where you can ship a bespoke fix on Tuesday because a customer needs it on Wednesday - and where that work is recognised as the most valuable engineering at the company. Loman is that company.