> Markdown version of [/jobs/ext/4010617-forward-deployed-senior-engineer-ai-analytics](https://www.wearedevelopers.com/jobs/ext/4010617-forward-deployed-senior-engineer-ai-analytics). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Forward Deployed Senior Engineer, AI & Analytics - **Company:** Worldwide Flight Services - **Location:** United States (Remote available) - **Experience:** Expert - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Continuous Integration, Customer Data Management, Python (Programming Language), Machine Learning, Operational Data Store, Cloud Services, Design of Telemetry/logging Agents, Retrieval-Augmented Generation, Large Language Models, Backend, Fastapi, Kubernetes, Information Technology, Production Code, Data Analytics, Machine Learning Operations, Api Design, Human in the Loop, Docker - **Published:** October 29, 2026 - **Apply:** https://arc.dev/remote-jobs/j/redirect/pvbhrj8pi6 ## About the Role * 6+ years in software or ML engineering, including customer-facing or business-embedded delivery. * Track record of scoping and shipping complex systems end to end in ambiguous, fast-moving environments, and staying accountable after go-live. * Production experience with LLM or ML systems: prompt and agent design, RAG, evaluation frameworks, deployment at scale, and an understanding of how model behaviour shapes product experience. * Strong Python and API development (FastAPI or equivalent); comfortable across backend and front-end enough to build a working system alone. * Practical depth in Docker, Kubernetes, CI/CD, and cloud data platforms; batch and real-time integration against messy enterprise sources. * Fluency in the analytics half of the job: semantic and dimensional modelling, modern BI, and designing data products that non-technical and external users can operate without help. * Aviation, logistics or complex operations exposure is an advantage, not a prerequisite. Willingness to learn the operation from the floor is not optional. Qualifications * Bachelor's, MSc or PhD in Computer Science, Engineering, Mathematics or related field. Personal characteristics * Outcome ownership, measured on whether the operation changed, not on assets produced. * Structured problem solving under ambiguity; able to decide with incomplete information and adjust. * Credibility with both a ramp supervisor and a platform architect; clear written and verbal communication. * Calm under operational pressure. Go-lives are rarely quiet. ## Description Deploy and land * Own end-to-end delivery of AI and analytics products at stations in your region: discovery, scoping, build, integration, UAT, hypercare and rollout. * Embed with the operation. Sit in business stand-ups, walk the ramp and the cargo floor, and specify the local data reality rather than assuming it. * Ship production code in Python and comparable stacks against live flight, cargo, workforce and equipment data, not prototypes handed to someone else to productionise. * Scale deployment. Reduce cost and time-to-deploy from station N to station N+1 through reusable integrations, configuration and runbooks. Build on the platform, not around it * Consume through the semantic layer and the governed data platform. No parallel pipelines, no local definitions, no workarounds. * Apply CoE patterns and guardrails: reference architectures, AI registry, human-in-the-loop controls, access and residency rules. * Integrate GenAI and ML into workflows: RAG and retrieval, agentic and tool-use patterns, evaluation frameworks, observability and cost control in production. Make insight self-service * Build data products, not report queues. Design- governed, reusable datasets and semantic models that let operations, airlines and forwarders answer their own questions. * Put insight in front of the customer. Deliver station and customer-facing analytics, including natural-language access to operational data, with the access, residency and confidentiality controls customer data demands. * Design for the user, not the requester. Every product names the decision it changes and who makes it. Instrument usage, retire what nobody opens, consolidate the long tail rather than extend it. Close the loop * Feed the field back. Every integration problem you solve becomes a hardened pattern for the CoE and a requirement for the platform team. * Own adoption and benefits. Instrument what you deploy, track usage and benefit against the business case, and report it honestly. * Be the technical face to the business and to customers, translating between operations, the CoE and platform engineering. Success in the First 12 Months Measure What Good Looks Like Stations live Flagship AI and analytics products deployed and running at named sites in region Adoption Sustained use by the operational teams and customers the product was built for Benefit realized Measured movement against the business case, evidenced not asserted Self-service Users answering their own questions; standing report requests reduced