Principal Solutions Architect

Job description

## **Company Overview:** [Maven AGI is an enterprise](https://www.mavenagi.com/) AI platform on a mission to unleash business artificial general intelligence (AGI), starting with customer service. Founded in July 2023 by executives from HubSpot, Google and Stripe, Maven builds conversational AI agents capable of delivering accurate, autonomous support that delights customers at scale. Our platform unifies fragmented systems, integrates knowledge and personalization sources, and enables intelligent actions - all without costly system changes. We’re laying the foundation for a future where our technology handles complex tasks, allowing people to focus on what they do best: creative problem-solving, relationship building, and delivering exceptional customer experiences. We’ve started by reimagining the enterprise customer experience with a support use case. We believe that today’s support experience is broken: slow and painful for customers, and expensive and human capital intensive for companies. We are building Maven to deliver better, cheaper support, for both end users and agents. With recent advancements in Generative AI, it is now possible to deliver delightful customer experiences at a fraction of today’s cost. # **Role summary** Maven is scaling from bespoke AI agent deployments to repeatable enterprise implementations. Our largest customers need someone who can de-risk complex technical architectures before signature, guide production launches across messy enterprise systems, and turn each deployment into reusable leverage for the next one. As a **Principal Solutions Architect**, you will be the technical design authority for Maven’s most strategic enterprise accounts. You will own pre-SOW technical validation, design production-grade AI agent architectures, guide Forward Deployed Engineers from POC through launch, and help convert field learnings into reusable platform capabilities. This is a forward-deployed, customer-facing technical leadership role. You should be comfortable prototyping quickly, validating quality with data, explaining complex AI systems to executives, and going deep with staff engineers on integrations, data flows, evals, observability, privacy, and fallback behavior. ### **What you’ll own** **Pre-SOW technical validation** Partner with Sales, Engagement, and Solutions Engineering to validate platform fit, integration complexity, data/security constraints, FDE effort, and implementation risks before strategic SOWs are finalized. **Production architecture for strategic accounts** Design reference architectures for multi-channel AI agent systems across voice, chat, email, and SMS, including integrations with CCaaS, CRM, data warehouse, identity, and internal knowledge systems. Define data contracts, tool/action patterns, observability, privacy boundaries, fallback behavior, and acceptance gates. **Technical direction from POC to launch** Guide FDEs through architecture decisions, edge cases, CI/CD, evals, monitoring, rollback plans, and launch readiness. You own technical direction and architecture quality; FDEs own implementation. **Reusable deployment leverage** Turn each strategic deployment into reusable assets: reference architectures, integration templates, eval harnesses, implementation checklists, demo baselines, or product requirements that reduce effort for future launches. **Executive and technical customer communication** Lead architecture reviews, security reviews, roadmap sessions, and technical deep dives with VP/C-level sponsors, security teams, and staff engineers. Make trade-offs clear for both executive and deeply technical audiences. **Field-to-product feedback loop** Identify repeated customer needs and implementation gaps. Partner with Product and Engineering to turn field patterns into roadmap inputs, reusable platform capabilities, and core product improvements. # **What success looks like** Within your first 6–12 months: - Strategic deals have clearer technical feasibility, integration scope, risk, and effort estimates before SOW. - FDE teams move faster because they have stronger architecture guidance, launch gates, and reusable implementation assets. - Complex deployments launch with better observability, evals, fallbacks, and privacy controls. - Customers trust Maven’s technical direction in security, architecture, and executive-level reviews. - Repeated deployment patterns become productized rather than remaining bespoke. # **What makes you a fit** ## **Required** - 7+ years in Solutions Architecture, Applied Engineering, Forward Deployed Engineering, Sales Engineering, or equivalent customer-facing technical leadership for enterprise SaaS, data, ML, or AI products. - Track record taking complex enterprise technical solutions from discovery or POC into production. - Strong system design skills across APIs, event-driven systems, data modeling, identity/consent flows, latency, throughput, reliability, and observability. - Practical AI/LLM depth: RAG, tool use/function calling, prompt and retrieval evaluation, safety guardrails, agent orchestration, and quality measurement. - Hands-on technical fluency in TypeScript/Node, Python, or similar. You are not the primary production engineer, but you can prototype, review code, debug integrations, and reason through implementation trade-offs. - Excellent written, verbal, and visual communication. You can lead a security review, produce a clear architecture diagram, and present trade-offs to executives. - Builder mindset. You want to help create the Solutions Architecture function, not just operate inside an existing one. ## **Nice to have** - Contact center / CCaaS experience with platforms like Genesys, Five9, NICE, Twilio, or Amazon Connect. - Depth with Zendesk, Salesforce Service Cloud, ServiceNow, or similar customer support systems. - Experience in regulated or complex data environments such as financial services, healthcare, gaming, marketplaces, or franchise businesses. - Experience at a frontier AI company, FDE-style company, data/ML platform company, high-growth SaaS company, or as a technical lead on complex enterprise AI programs. - Prior experience building a technical function from zero to one.