Forward Deployed AI Product Engineer

phData

2d•Remote

About The Position

The Forward Deployed AI Product Engineer (Vector) is a customer-embedded, full-stack engineer who builds AI-native applications and agents that solve high-value business problems on top of the client’s data and AI platform. This role sits at the intersection of product, engineering, and business outcomes: you will live inside a business unit, learn its operations in detail, and rapidly ship working software that proves measurable ROI. You will be deployed directly into strategic customer accounts as the tip of the spear for AI-powered solutions. You will partner closely with business stakeholders, central platform teams, and other phData consultants to: Identify and shape AI use cases that were previously considered too messy, too manual, or too cross-functional to tackle. Rapidly prototype full-stack, MVP-grade agents, applications, and workflows that demonstrate value in weeks, not quarters. Productionize the most impactful solutions, integrating with the client’s AI Data Platform, existing systems, and operational workflows. Feed technical and architectural patterns back into the client’s platform and operating model to create durable, long-term leverage. You will own the end-to-end lifecycle of AI-native solutions in your business area: discovery, value articulation, design, implementation, iteration, and handoff. You must be equally comfortable whiteboarding with executives about KPIs and ROI, sitting side-by-side with frontline operators to understand real-world constraints, and diving deep into code to ship features quickly. This is a highly hands-on, travel-heavy role designed for engineers who want to be as close as possible to the business problem, not just the technology. You will act as a lead engineer within one or more “fusion pods,” collaborating with data engineers, ML engineers, activation specialists, and client teams to deliver outsized business impact.

Requirements

You are obsessed with business outcomes, not just technology. You think in terms of KPIs, ROI, and time-to-value, and you can sit across from a VP, GM, or CTO to recommend architectures, workflows, and tools that directly tie to their goals.
You are a builder. You’re not looking for a rigid playbook or a fully defined product; you want to help define the solution, build the first versions, and iterate quickly based on real user feedback.
You have strong business curiosity. You naturally dig into how operations actually work, what’s broken, and how AI and software can reshape processes rather than just automate existing ones.
You have empathy to sit with business users, observe how they actually work, and listen carefully to their pain points. You apply design thinking to reimagine their workflows and tools, not just digitize the current process.
You are comfortable working in ambiguity. The problems you tackle are often poorly defined at the start; you help clarify the problem, design the solution, and drive execution.
You have shipped production-grade applications that connect powerful data and AI back ends with intuitive front-end experiences (web apps, internal tools, or operational interfaces).
You are fluent in AI-native development patterns: RAG, LLM orchestration, agentic workflows, and modern AI evaluation and guardrails. You see AI tools as essential infrastructure for modern engineering.
You can rapidly prototype with modern tools (Streamlit, Gradio, lightweight React apps, workflow tools, agents) and are comfortable throwing away code to converge on the right solution.
You are a strong, opinionated communicator. You can explain complex technical concepts in simple language, lead client conversations, and build trust with both executives and hands-on operators.
You are willing and excited to travel frequently and spend significant time on-site with clients to build deep operational empathy and accelerate delivery.
You are energized by turning one-off successes into repeatable patterns: documenting what works, identifying platform gaps, and helping shape the client’s Intelligence Platform roadmap.
You are prepared to step into what is likely to be one of the most demanding roles of your career, and you are motivated by the impact, autonomy, and responsibility that comes with it.
Frontend: Streamlit, Gradio, React, or similar.
Databases and Storage: Postgres, cloud-native warehouses, Vector DBs, Graph DBs.
Workflow and Automation: Cloud orchestration tools, CI/CD pipelines, and AI-native workflow systems; familiarity with tools like Microsoft Copilot or similar is a plus.
Platforms: Snowflake and major cloud providers (AWS, Azure, GCP); experience with modern data and AI platforms and APIs.
Agentic AI Skills: Practical experience with RAG, LLM workflows, agentic frameworks, prompt engineering, and implementing guardrails and evaluations for AI systems.
Curious and Continuous Learner: Demonstrated ability to upskill rapidly on new APIs, data platforms, workforce systems, and emerging AI tooling.
BA/BS in computer science, engineering, mathematics, operations, or related fields preferred. Equivalent practical experience considered.
4+ years of hands-on software engineering experience (more for Senior/Principal levels), including building and operating production systems.
Travel up to 50% (client-dependent). Frequent on-site presence with customers is expected for this role.

Nice To Haves

Prior consulting, customer-embedded engineering, or product engineering experience in data/AI-heavy environments strongly preferred.

Responsibilities

Rapid Prototyping: Design and build full-stack, MVP-grade prototypes for AI-native applications, agents, and workflows that can be validated quickly with end users.
Full-Stack Execution: Deliver end-to-end solutions using a mix of Python, SQL, and TypeScript/JavaScript, including APIs, back-end services, and front-end experiences.
AI Agents and Workflows: Implement AI-powered workflows and agents that solve real business problems, including RAG pipelines, AI agents, workflows, skills, memory, and model-context protocols where appropriate.
User Interfaces: Build high-fidelity, production-ready user interfaces (e.g., Streamlit/Gradio/React or similar) that make AI capabilities usable and intuitive for business users.
Data Transformation/Manipulation: Move, transform, and harmonize disparate data sources into the client’s data platform (e.g., Snowflake, Databricks, cloud data warehouses), exposing clean, reliable data to applications and agents.
Pattern Feeding: Capture and contribute technical patterns, best practices, and anti-patterns back into the client’s platform and phData’s internal playbooks to build long-term leverage.
Platform Advocacy: Bridge the gap between business-unit needs and central platform capabilities. Identify capability gaps (CI/CD, orchestration, observability, streaming ingestion, vector search, AI gateways, etc.) and help prioritize their development.
Pull-Through Identification: Identify new data sources, platform capabilities, and integration opportunities that create demand for additional phData services (data engineering, ML engineering, Analytics, Workforce AI, platform operations).
Use Case Discovery: Lead or support discovery sessions with business stakeholders to surface high-impact AI opportunities, unconstrained by legacy processes or “the way we’ve always done it.”
Value Articulation: Define success metrics in terms of business KPIs and quantify measurable ROI to justify additional services and platform investments.
Operational Empathy: Spend significant time on-site with end users and operational teams to deeply understand workflows, constraints, and edge cases; design solutions that fit real-world usage.
Client-Facing Communication: Engage confidently with executives, managers, and technical teams. Clearly communicate trade-offs, timelines, and risks, and maintain trust while moving quickly.
Documentation and Enablement: Contribute to internal and client-facing documentation, run-throughs, demos, and training that help others adopt and extend the solutions you build.