AI Forward Deployed Engineer

This is a fully remote position, open to applicants in India.

📋 Description

• Take ownership of AI deployments from start to finish.

• Lead the technical onboarding process for AI Agents, Copilot, and various AI functionalities.

• Set up knowledge sources, workflows, automation rules, and AI behaviors.

• Integrate customer systems to enable production-ready deployments.

• Collaborate with customer stakeholders to define rollout plans and success criteria.

• Continually enhance deployment playbooks and best practices for implementation.

• Serve as the first point of technical investigation.

• When customers inquire, *"Why did the AI respond this way?"* You will:

• Analyze AI behavior by utilizing logs, traces, evaluation tools, and retrieval diagnostics.

• Determine if issues arise from knowledge quality, retrieval failures, prompt or workflow configurations, or product limitations.

• Resolve issues through configuration whenever feasible.

• Escalate to engineering only when presenting well-diagnosed, reproducible problems.

• Create custom solutions.

• Develop production-quality integrations, scripts, and automations.

• Build connectors using customer APIs.

• Conduct data migrations and transform knowledge bases.

• Prototype solutions tailored to customer needs where product capabilities are lacking.

• Transform recurring workarounds into proposals for product enhancements.

• Enhance AI quality.

• Conduct structured AI quality reviews using established evaluation frameworks.

• Measure and improve response quality, deflection, and resolution rates.

• Adjust AI behavior through knowledge enhancements, workflows, and automation logic.

• Substitute anecdotal feedback with quantifiable quality metrics.

• Influence product development.

• Being closest to customer deployments, you'll help shape the product roadmap by:

• Identifying recurring deployment challenges.

• Monitoring feature gaps and customer pain points.

• Providing structured field insights to Product and Engineering teams.

• Differentiating between product improvements and implementation best practices.

• Empower internal teams.

• Develop deployment playbooks and troubleshooting manuals.

• Train Customer Success teams on AI administration and diagnostics.

• Minimize reliance on Product and Engineering for routine customer issues.

• Support scaling AI deployments through comprehensive documentation and operational excellence.

⛳️ Requirements

• Strong foundation in full-stack engineering — 3 to 6 years of software engineering experience with the ability to deliver production-quality code independently (Python + TypeScript/JS is likely a good fit for our stack). Not merely a scripter; a true engineer.

• Experience with APIs and integrations — comfortable interpreting customer API documentation, creating connectors, and transferring data between systems (such as email systems, helpdesks, CRMs, and webhooks).

• Familiarity with LLM systems — understands RAG pipelines, retrieval failure modes, prompt/instruction design, vector search, and the reasons behind incorrect AI responses. Hands-on experience with evaluation/observability tools (we utilize Langfuse-style tracing, LLM-as-judge, golden datasets — candidates should be able to read and extend these tools).

• Proficient data skills — strong SQL capabilities, comfortable handling messy customer data, and able to conduct log analysis at scale.

• Production debugging ability — capable of methodically investigating live issues: reproducing, isolating, root-cause analyzing, and documenting.

• Basic cloud/infrastructure knowledge — sufficient understanding of AWS/GCP to grasp deployment constraints, authentication (OAuth/SSO), and data security inquiries from customers.

• Composure in customer-facing situations — able to manage a call with a frustrated support ops leader, set honest expectations, and leave them feeling more confident.

• Two-way translation skills — capable of converting vague customer complaints into precise technical diagnoses and translating technical constraints into clear business language.

• Ownership in ambiguous situations — thrives under incomplete requirements; scopes a minimum viable product fix, implements it, and iterates. Does not wait for a ticket to be fully specified.

• Sound judgment on escalation — distinguishes between "I can resolve this with configuration" and "this is a product gap that engineering must address" without raising unnecessary alarms.

• Discipline in documentation — playbooks, runbooks, and field reports are crucial to the role. If it isn't documented, the time savings won't accumulate.

• Ability to prioritize across multiple accounts — able to manage multiple deployments and queues without a project manager directing traffic.

• Must maintain work-hour overlap with Engineering, US-based teams, and customers to encourage effective cross-functional collaboration.

🏝️ Benefits

• Competitive salary and performance-based bonuses.

• Comprehensive benefits package including health, dental, and vision coverage.

• Opportunities for professional growth and development.

• Flexible working hours and remote work options.

• Collaborative and inclusive work environment.