As an AI Developer on Xplore's Data & AI team, you will design, build, and deploy the agentic workflows and AI-enabled applications that change how the business runs - across network operations, customer experience, field services, and back-office functions.
There are more problems worth solving here than any single roadmap can hold. That is why this role is built for a systems thinker. You will need to see how systems connect: where a process actually breaks versus where it merely looks broken, which upstream system owns the truth, what a change does three steps downstream, and which piece of scaffolding built today makes the next five workflows cheaper to ship. You will be expected to bring a point of view on sequencing, not just execution.
Unlike a research-oriented ML engineer focused on model development, or a software engineer focused on a single application surface, your specialty is orchestration and integration: decomposing messy, human-shaped processes into reliable, observable, governed agentic systems.
Key Responsibilities Include:- Partner with process owners across operations, engineering, and corporate functions to map existing workflows end to end, identify high-value automation candidates, and distinguish what should be automated from what should first be redesigned or retired.
- Design, build, and deploy production agentic workflows - tool use, retrieval, multi-step orchestration, and human-in-the-loop checkpoints - that execute real business processes rather than demonstrate them.
- Integrate agentic capabilities into existing and new web applications, building the services, APIs, and user-facing touchpoints through which people interact with agents.
- Define what "working" looks like with stakeholders before building: acceptance criteria, quality thresholds, and measurable process outcomes such as cycle time, human touch rate, error rate, and cost per transaction.
- Own the reliability of everything you deploy: evaluation harnesses, regression testing against real cases, monitoring and tracing, known failure modes, graceful degradation, and clear escalation paths back to humans.
- Establish and enforce guardrails for AI systems in production, including least-privilege data and tool access, sensitive data handling, prompt injection and tool-abuse mitigation, and full auditability of agent actions.
- Build reusable platform components - shared connector and tool libraries, orchestration patterns, prompt and evaluation scaffolding - so each new workflow ships faster and safer than the last.
- Produce architecture documentation, decision records, and operational runbooks so owned systems can be understood, operated, and extended by others.
- Advise leadership on AI opportunity sizing and sequencing, including honest assessment of what is not yet ready for production and why.
The ideal candidate will possess:- 5+ years of software engineering, data engineering, or applied AI experience, with recent hands-on delivery of production systems.
- Demonstrated experience taking LLM-based or agentic applications into production with real users - beyond prototypes and proofs of concept.
- Strong Python skills, including API design, asynchronous patterns, and service architecture.
- Practical fluency with agentic building blocks: tool and function calling, retrieval, context management, state and memory, multi-step orchestration, and the tradeoffs between competing approaches.
- Sufficient full-stack capability to integrate AI into web applications: REST or GraphQL API design, authentication and authorization, and working competence with a modern front-end framework such as React.
- Systems thinking demonstrated in practice: the ability to reason about coupling, failure propagation, and second-order effects across connected systems, and to defend architectural tradeoffs to both engineers and executives.
- Proven ability to work directly with non-technical stakeholders to elicit requirements from ambiguous, undocumented, or contested processes.
- Sound judgment on security and data handling in AI systems, including access scoping, secrets management, and sensitive data classification.
- Excellent written and verbal communication skills; comfort presenting technical decisions and tradeoffs to Director and VP-level audiences.
Preferred Qualifications- Experience with agent evaluation and observability tooling, including tracing, offline and online evaluation, and structured human review.
- Experience implementing AI governance, acceptable-use, or model risk frameworks in an enterprise or regulated environment.
- Familiarity with emerging tool-integration standards such as the Model Context Protocol, and with connecting agents to enterprise systems of record.
- Background in telecommunications, network operations, or infrastructure environments.
- Experience with cloud platforms and containerized deployment, and with CI/CD for AI-enabled services.
- Experience working against a governed enterprise data platform such as Databricks, including catalog-based access controls.
- Bachelor's degree in Computer Science, Engineering, Information Systems, or a related field, or equivalent practical experience.
