Software Engineer, Applied AI

What we do

• Build the insurance behind the physical world.
• The world is building data centers, energy projects, hospitals, schools, and critical infrastructure. Every one of those projects needs insurance to move forward. Yet the process of getting that insurance still depends on scattered documents, manual work, and weeks of back-and-forth.
• Shepherd is rebuilding commercial insurance with software, data, AI, and deep underwriting expertise. We started in construction and energy, where understanding the specifics of a project matters and better decisions have consequences far beyond a screen.
• We develop insurance products, decide which risks to insure and how to price them, and partner with established insurers to back the coverage. We own the platform our team uses to turn complex submissions into coverage for the builders and operators shaping the world around us.
• Real problems. Real ownership.
• Our engineers, designers, underwriters, and actuaries build together. The problems span turning messy documents into reliable data, making pricing more precise, and designing AI agents that can take on increasingly complex underwriting work.
• The ambition is a platform where experts spend more time on judgment and relationships, supported by software that handles the work around them. Join us to build systems whose decisions help real projects get built.
• 630+ customers / $500B+ insured project value
• Our investors
• Capital to build. Partners to grow with.
• $67M total funding / $42M Series B in March 2026
• Shepherd is backed by Intact Private Capital, Costanoa Ventures, Spark Capital, Susa Ventures, and Y Combinator. In March 2026, we announced a $42M Series B led by Intact Private Capital, bringing our total funding to $67M.
• Intact is also a capacity partner. Alongside its investment, Intact works with Shepherd to bring insurance products to market. That relationship connects our platform and underwriting expertise with the resources of a global insurer.
• A bigger opportunity ahead

We are investing in new insurance products, our underwriting team, and the path toward autonomous underwriting. For the people joining Shepherd, that means the chance to shape both the technology and the business it makes possible.

Our backing brings together investors who build technology companies and partners who understand insurance. We are building a company that earns the confidence of both.

The Role

We are hiring a Software Engineer focused on Applied AI to build intelligent systems that directly power our underwriting and pricing engine. This role sits at the intersection of product engineering, AI infrastructure, and real-world insurance workflows.

You will design and ship production AI systems that meaningfully improve how we evaluate risk, price policies, and operate our business.

What You’ll Do

• Partner with product, engineering, underwriting, and data teams to identify high-leverage AI opportunities
• Design, build, and deploy LLM-powered features across underwriting, pricing, and internal tooling
• Develop AI agents and retrieval-augmented systems that extract structured insights from unstructured insurance data
• Build production-grade AI infrastructure for model serving, inference, evaluation, and monitoring
• Create internal AI tools that enable underwriters and operators to move faster with higher quality decisions
• Establish best practices for prompt engineering, evaluation frameworks, and model reliability
• Own AI features end-to-end from ideation to production

What We’re Looking For

• Strong backend engineering experience building scalable APIs and distributed systems
• Hands-on experience shipping AI-powered products into production
• Familiarity with LLMs, RAG architectures, embeddings, and vector databases
• Experience integrating AI systems into user-facing or workflow-driven applications
• Comfort with cloud infrastructure and deploying AI workloads at scale
• A product mindset and the ability to translate ambiguous business problems into concrete AI systems
• You have at least 5+ years of working experience (preferably as a full-stack engineer) with a track record of delivering high-impact products
• Bonus Points
• Experience building AI agents or multi-step reasoning systems
• Background in applied machine learning beyond generative AI
• Experience with pricing systems, financial infrastructure, or risk modeling
• Interest in transforming legacy industries with modern AI systems
• Why This Role Matters
• Insurance is one of the largest industries in the world, yet much of underwriting and pricing remains manual and fragmented. At Shepherd, AI is not a side project – it is core to how we build durable competitive advantage.
• You will help define how applied AI transforms risk evaluation in high-hazard industries and shape the future of AI-native insurance infrastructure.
• Our Tech Stack
• Frontend
• React & Next.js
• Typescript
• GraphQL w/ Apollo
• Backend
• Node.js
• Postgres & Redis
• Typescript
• GraphQL APIs
• How we work
• Shepherd runs on four values. Here's what each one means in this seat.
• Think big, build big.
• We exist to protect progress and the industries that rely on it. The work here is aimed at a system that runs on its own, and the roadmap gets sequenced backward from that rather than forward from what's easy.
• Win together.
• We rise as one. We support each other, raise the bar, and celebrate collective success. Come set a standard the rest of the team inherits.
• Cross the aisle.
• Collaboration wins. We listen deeply, work across boundaries, and prioritize shared success over individual lanes. The best product calls here come from engineers who've sat with underwriters and underwriters who understand where the model breaks, and much of this job is listening closely enough on both sides to make that happen.
• Go get it.
• We act with urgency, move with confidence, take smart risks, and push forward with intention. Nobody hands you the roadmap, the data, or the meeting invite. You pull the failing runs, book the time with the underwriters, and decide what matters.