Staff Engineer (Applied AI / ML)

What you'll do

1. AI Discovery & Technical Product Leadership
• Stakeholder Ideation: Support client discovery workshops, whiteboarding sessions, and interviews with client business and product owners to identify operational bottlenecks suitable for AI intervention. Successful candidates will be comfortable engaging directly and frequently with the client, demoing work early and often, gathering feedback, and closing tight iterative loops throughout the development process. 
• Use Case Prioritization & ROI Analysis: Evaluate potential AI use cases across business impact, technical viability, data readiness, and ROI to build prioritized product roadmaps and PoC scopes. Recommend pragmatic alternatives when the requested approach is unlikely to be the simplest, highest-value, or most effective solution. 
• Requirements to Spec Translation: Bridge the gap between vague business asks and clear technical architecture, translating functional requirements into PRDs, technical specifications, and verifiable prompt/eval criteria. Constructively challenge proposed solutions when appropriate and recommend simpler or more effective technical approaches. 
• Rapid Prototyping (0-to-1): Rapidly build and demonstrate working proof-of-concept (PoC) apps and interactive demos (within 2–3 week cycles) to validate value with local executive stakeholders before full investment.
• Prototype-to-Production Hardening (1-to-N): Evolve successful prototypes into reliable, secure, maintainable production systems by strengthening architecture, testing, observability, scalability, evaluation, and operational readiness while preserving delivery velocity. 
2. Architecture & Core AI Engineering
• Agentic Systems & Harness Design: Design and build robust single- and multi-agent systems, including tool use, control loops, context and state management, planning, delegation, structured outputs, and failure recovery. Build the harnesses and runtime primitives that make agent behavior reliable, inspectable, and maintainable.
• Full-Stack Product Engineering: Work comfortably across the stack - from front-end experiences and backend services to APIs, data infrastructure, and cloud deployment - to get software shipped. You do not need to be a specialist at every layer, but you should bring strong software engineering fundamentals, learn quickly, and be willing to work wherever the product needs you most. 
• Retrieval & Enterprise Integration: Build retrieval and grounding capabilities where appropriate-including semantic and hybrid search, reranking, metadata filtering, and RAG- - and integrate AI systems with enterprise APIs, data platforms, workflows, and user-facing applications.
• Machine Learning Foundations & Model Adaptation: Bring a strong grounding in machine learning fundamentals. While most initiatives begin with frontier or open-source pretrained models, understand when fine-tuning is warranted and how to apply techniques such as LoRA/PEFT, dataset preparation, training pipelines, and model evaluation.
 
3. LLMOps, Governance & Production Hardening
• AI Evaluation, Observability & Quality: Build rigorous evaluation and observability frameworks for probabilistic AI systems, combining offline benchmarks, task-specific metrics, LLM-as-judge, human evaluation, regression testing, and production telemetry. Measure quality, latency, cost, and failure modes, then use those signals to systematically improve models, prompts, tools, retrieval, and system architecture.
• Safety, Guardrails & Reliability: Design enterprise-grade protections including moderation, PII handling, prompt-injection defenses, permission-aware tool use, fallback logic, and graceful failure modes to keep AI systems safe, predictable, and resilient in production.

What you'll bring

• Experience: 7+ years in software engineering or technical product engineering, with 1+ years dedicated to building and deploying Generative AI / LLM applications in production.
• Curiosity, Judgment & Consulting Poise: Bring intellectual curiosity, humility, and a commitment to first-principles thinking. We want people who actively shape solutions with the business, challenge assumptions constructively, and redesign the work when needed - while doing so with the empathy, communication skills, and stakeholder awareness expected of a trusted consultant.
• Product & Business-Facing Acumen: Demonstrated success in client-facing or internal product leadership roles (e.g., Forward-Deployed Engineer, AI Solutions Architect, Technical Product Manager). Proven ability to facilitate discovery workshops, extract requirements, and pitch technical solutions to executive audiences.
• Programming & API Architecture: Expert proficiency in Python and strong software engineering fundamentals, with experience designing modern stateless and stateful services for real-time, interactive, and batch workloads. Comfortable working across protocols and integration patterns such as REST, WebSockets, MCP, streaming APIs, and event-driven architectures.
• Spec-Driven & Agentic Engineering Tools: Hands-on experience utilizing spec-driven AI development tools and agentic IDEs (e.g., Cursor, Codex, OpenCode, etc) to write structured software specifications, orchestrate autonomous code generation, and execute end-to-end task flows. 
• AI Orchestration & Harness Design: Understand modern agent frameworks and when to use - or avoid - them. Be comfortable building custom harnesses and control loops when greater flexibility, reliability, or control is needed.
• Retrieval Architecture & Agentic Search: Deep expertise designing and optimizing retrieval systems, including embedding search, hybrid retrieval, metadata filtering, reranking, and agentic search patterns. Know how to measure retrieval quality, diagnose failure modes, and systematically tune relevance, recall, precision, latency, and cost.
• Cloud & DevOps: Solid experience with cloud infrastructure (AWS, GCP, or Azure), containerization (Docker, Kubernetes), and CI/CD automation.
• erfaces, integration mechanisms (e.g., async vs. synchronous), middleware, etc.
• Experience with typical AuthN/AuthZ methods and products
• Experience managing data, including the selection of persistence product, design of database schema, constraints and transaction boundaries, read/write design trade-off decisions and how it relates to mutable vs. immutable data state
• Experience balancing feature development vs. technical debt accumulation in order to deliver business needs while also maintaining quality over time
• Bachelor's degree in engineering, information systems, computer science, business administration, or other related fields

Preferred Experience

Experience working on-site within high-velocity consultant/engineer delivery pods or client innovation labs.
Experience deploying, operating, and supporting AI/ML systems in production within large-scale enterprise environments, including navigating the reliability, security, governance, and integration requirements that come with them.
Prior exposure to Media & Entertainment (M&E), Digital Media Supply Chain, or enterprise content automation.
Experience working with video, imagery, audio, and other unstructured media, including applying AI models to understand, search, classify, transform, or generate insights from multimodal content. 
Hands-on experience rapidly vibe-coding functional prototypes and UI’s with modern coding agents and agentic harnesses such as GitHub Copilot, Claude Code, and OpenCode. Able to turn ideas into working demos quickly while knowing when prototype-grade shortcuts need to be replaced with production engineering.

Travel Requirements