Founding Senior Manager of Forward Deployed Engineering

Afresh Technologies; San Francisco (CA); On-site; USD 200,000 - 300,000; Full time

Benefits offered by this job

Health plans

Parental leave

Equity packages

401k matching

Flexible vacation

Professional development

Work from home stipend

Job summary

Afresh Technologies is seeking a player-coach Architect to build and lead the forward deployed engineering team embedded with large grocery customers. You’ll own the technical relationship with the customers’ engineering leaders, review architectures, and steer execution across multiple accounts while staying hands-on when needed.

This role sits in an existing AI engineering group but will be built from the ground up, with accountability for hiring, standards, and the platform’s evolution.

Qualifications

• 10+ years in software engineering, solutions architecture, consulting or a technical customer-facing role.
• 2+ years managing engineers in services or forward deployed settings.
• Own architecture across accounts and lead multi-account deployments.
• Willingness to travel 25–40% to customer sites.
• Grocery, retail or supply chain data experience is a plus.

Responsibilities

• Lead the forward deployed engineering team and set technical standards.
• Review architectures, unblock the team, and contribute code when needed.
• Manage concurrent engagements and decide what to build and in what order.
• Hire, develop, and retain engineers; define reference architectures.
• Partner with our AI engineering group and interface with enterprise data leadership.

Skills

Software engineering

Solutions architecture

Customer-facing

Consulting

Tools

Databricks

BigQuery

Snowflake

Job description

• Afresh is building the team that deploys our AI inside the largest grocers in the US
• You’ll build and lead the team of forward deployed engineers who embed with our enterprise grocery customers, integrating into their data, shipping AI systems on top of it, and hardening what works into our platform
• You’ll own the technical relationship with each customer’s engineering leaders, and you’ll partner with the Deployment Strategist who owns the account to plan and sequence the work
• This is a player-coach role with a strong bias toward leadership
• You’ll review architecture, unblock your engineers, and get into the code when an account needs it, but most of your week goes to leading the team’s execution and to customer stakeholders rather than to personally building
• This team will be part of an existing AI engineering group, but will be built and scaled from the ground up
• We’re looking for an architect rather than a specialist
• You should be able to hold a grocer’s data estate and our platform in view at the same time and judge where the two meet, without needing to be the deepest expert in any single layer
• Lead the team
• Hire, develop and retain forward deployed engineers. Set the technical bar and the standard for how the team works with customers
• Lead execution across concurrent customer engagements: what gets built, in what order, and to what quality bar
• Staff engagements across accounts, making resourcing calls that balance customer need, team development and what the platform needs next
• Run regular 1:1s and invest in career growth. Several of our engineers are strong builders doing customer-facing work for the first time
• Stay hands-on enough to review architectures and code, lead technical discovery sessions, and debug alongside your team when an account calls for it
• Own the technical relationship
• Be the senior technical partner to each customer’s engineering and data leadership. Set the delivery roadmap with them, defend the architecture behind it, and tell them early when something slips
• Partner with the Deployment Strategist on overall account execution. They own the customer relationship and the commercial picture; you own the technical plan and whether it lands
• Turn a broad business goal, like reducing shrink, into a scoped program with a sequence, a cost and a date
• Work out how each customer actually operates, most of which is undocumented: what they mean by an “item,” who signs off on a markdown, which exception path the DC team uses every week. The real constraint is usually not the stated objection
• Own the architecture across accounts
• Own one reference architecture for AIE that holds across deployments, covering our current hosted default and the customer-cloud exceptions, so platform features ship once rather than once per customer
• Decide what gets hardened into the shared platform, what gets built for a single customer, and what we decline to build
• Set production standards for reliability, security review, data residency, and the evals and tracing that show whether a deployed AI system is actually working
• Close the loop into the platform
• Partner with our platform engineers to turn field learnings into product, so each deployment costs less than the one before it
• Build the repeatable playbooks, scoping templates and starter repos that capture what the team learns in the field
• Travel to customer sites roughly 25–40%, weighted toward engagement kickoffs and readouts
• We’re a mission-driven company that eliminates hundreds of millions of pounds of food waste in grocery stores every year, so your work has direct and visible impact
• You’d found the forward deployed team: its people, its standards and the way it works. That’s rare at a company already in production with enterprise customers
• This team reports into engineering, not sales. The job is deciding which customer problems are worth solving once for everyone, rather than closing the account in front of you
• You’ll stay technical. We want an architect who manages, not a manager who used to be technical
• Be part of an engineering culture that’s genuinely AI-forward — we want to be on the bleeding edge of agentic development, not watching from the sidelines
• Collaborative, supportive environment & awesome people

Benefits

• Comprehensive health plans
• Generous parental leave
• Equity packages
• 401k matching
• Flexible vacation policy
• Professional development program
• Work from home stipend

Executive presence, with the ability to move between a roadmap conversation with a customer’s VP of Data and a debugging session with your own engineerReal data fluency. You can open an unfamiliar enterprise data model and tell within a day which parts are trustworthy, and you plan around dirty data rather than being surprised by it mid-deployment. Fluent in SQL and a modern cloud data platform such as Databricks, BigQuery or SnowflakeProduction AI and LLM experience across retrieval, tool use and agentic workflows, with the habit of measuring quality rather than eyeballing it10+ years in software engineering, solutions architecture, consulting or a technical customer-facing role, including 2+ years managing engineers in a services, post-sales or forward deployed organizationArchitect’s range across ingestion, data modeling, application and AI, with the judgment to see how the pieces fit and where the risk sitsExperience delivering inside someone else’s enterprise: their cloud, their security review, their change management, their politicsWillingness to travel to customer sites roughly 25–40%Experience building a team from 0 to 1 rather than inheriting one: hiring, setting standards, and defining what good looks likeA track record of working out how a business actually runs when none of it is written down: what it treats as real, how work moves through it, which rules are policy and which are one person’s habitHaving built a forward deployed, professional services or solutions engineering function from scratchGrocery, retail or supply chain dataKnowledge graphs, ontologies or semantic layers running in productionDelivering the same product both vendor-hosted and inside a customer’s own cloud, and owning the abstraction between them