AI Integration Manager

This is a hands-on engineering role first. The AI Field Deployment Engineer (FDE) is responsible for the technical delivery of company's voice AI platform at enterprise scale in the US — owning solution architecture, system integration, cloud infrastructure, prompt engineering, and production operations for live customer deployments.

You are an engineer who deploys, debugs, and optimises production voice AI systems. You design the integration between the platform and a customer's telephony stack. You build the prompts that govern agent behaviour.

You own the cloud infrastructure that keeps systems running at SLA. You catch problems before customers do.

Client relationship and account growth are important — but they are the output of doing the engineering work exceptionally well, not a separate workstream.

Day one, you will take ownership of live enterprise deployments — including a large government services provider already in production — and ensure they are stable, optimised, and expanding. You will be companies technical ambassador in the US: the engineer enterprise CXOs and IT teams trust to deliver, and the voice that feeds real customer signal back into companyss product and engineering teams

This is a founding role. You will not inherit a playbook — you will write it.

Key Responsibilities

1. Technical Deployment & System Integration

 Own the end-to-end technical delivery of clients voice AI platform for enterprise clients — from solution architecture and integration design through development, UAT, go-live, and continuous optimisation.

 Lead integration with enterprise telephony infrastructure: SIP, RTP, call routing, signalling, and media handling across carriers, CPaaS providers, and on-premise telephony systems (Genesys, Avaya, Amazon Connect, Cisco).

 Design and implement integrations with enterprise backend systems — CRMs, ticketing platforms, data warehouses, and custom APIs — ensuring secure, scalable, and low-latency data flows.

 Write and maintain production code (primarily Python) for connectors, middleware, custom logic layers, and automation scripts that bridge clients platform with customer environments.

 Troubleshoot and resolve production issues across the full stack: ASR accuracy regressions, latency spikes, telephony signalling failures, API timeouts, and infrastructure incidents.

 Maintain complete system architecture documentation and own technical handoffs when deployments transition from build to steady-state operations.

2. Cloud Infrastructure & DevOps

 Deploy and manage companies platform components on cloud infrastructure (AWS / GCP / Azure) — provisioning, configuration, scaling, and cost management — ensuring deployments meet enterprise SLA requirements for uptime and performance.

 Containerise and orchestrate services using Docker and Kubernetes; manage Helm charts, deployments, and cluster operations in production environments.

 Build and maintain CI/CD pipelines to enable reliable, low-risk releases and configuration changes for

customer deployments.

 Design and implement monitoring, alerting, and observability stacks (e.g., Grafana, Prometheus, Datadog, CloudWatch) covering infrastructure health, model performance, and voice quality metrics.

 Build and maintain operational dashboards and runbooks that give both internal teams and customer stakeholders real-time visibility into deployment health.

 Own incident response for production environments — triage, root cause analysis, remediation, and post- incident review — with clear SLA commitments to enterprise customers.

3. Prompt Engineering & Voice AI Optimisation

 Design, iterate, and maintain the prompt architecture that governs companies voice agents across enterprise deployments — covering conversation flow, persona, intent handling, fallback strategies, and escalation logic.

 Systematically evaluate and improve agent performance using structured testing, A/B experimentation, and call transcript analysis — optimising for task completion rate, containment, latency, and customer experience metrics.

 Adapt prompts and agent configurations to customer-specific workflows, terminology, compliance constraints, and business rules — without compromising platform scalability.

 Stay current on developments in LLM orchestration, RAG pipelines, and agentic AI frameworks, and identify where they can improve production voice agent performance for companies customers.

 Build reusable prompt libraries, evaluation frameworks, and optimisation playbooks that scale across the growing number of US enterprise deployments.

4. Production Operations & Account Health

 Own production reliability and SLA performance for all live deployments — uptime, latency, call quality, and error rates — and take corrective action before customers escalate.

 Manage and grow existing enterprise accounts with a technical farmer's mindset: map new use cases, surface expansion opportunities, and work cross-functionally to translate those into deployed solutions and ARR growth.

 Conduct regular technical business reviews with customer stakeholders — tracking deployment health, demonstrating measurable ROI, and presenting a forward-looking roadmap for platform adoption.

 Serve as the primary technical point of contact for enterprise customers — trusted by their IT and engineering teams as the engineer who understands their environment as well as they do.

 Act as companies technical ambassador in the US market: credible at the engineering level, fluent at the executive level, and effective at building the relationships that make enterprises want to deepen their commitment to companies platform.

5. Internal Collaboration & Product Feedback

 Partner with companies product and engineering teams in India — translating field observations into structured product feedback, representing customer requirements in roadmap discussions, and stress-testing platform capabilities before they reach customers.

 Support pre-sales technical conversations, POC design, and solution architecture for new enterprise opportunities alongside the commercial team.

 Build and contribute to deployment playbooks, integration patterns, and technical runbooks that scale companies US delivery capability as the team grows.

Required Experience

Engineering depth is the primary filter. We are looking for a strong deployment engineer who also builds excellent

client relationships — not the other way around.

Engineering & Technical Skills (Primary)

 4–6 years of hands-on experience in a technical delivery role — solutions engineering, field engineering,

platform engineering, or technical consulting with real deployment ownership (not advisory-only).

 Proficiency in Python: able to write production-quality integration code, automation scripts, and data

pipelines — not just read and modify existing code.

 Cloud infrastructure experience on at least one major provider (AWS, GCP, or Azure): compute, networking,

storage, managed services, IAM, and cost management in production environments.

 Container and orchestration experience: Docker and Kubernetes in production — deployment management,

troubleshooting, and scaling.

 Hands-on experience with telephony and VoIP protocols: SIP, RTP, call routing, and media handling; working

knowledge of enterprise telephony or CPaaS platforms.

 Proficiency in prompt engineering: designing, testing, and iterating structured prompts for LLM-based or

voice AI systems — with a systematic, metrics-driven approach to optimisation.

 Experience building and operating monitoring and observability stacks: metrics, alerts, dashboards, and

incident response in production.

 Comfort operating in CI/CD environments: Git, pipelines, environment management, and release processes.

Client & Stakeholder Skills (Secondary, but real)

 Demonstrated ability to manage large enterprise accounts independently — multi-stakeholder navigation,

technical escalation management, and executive-level communication.

 Able to translate between engineering reality and business language fluently — in both directions.

 Track record of managing cross-functional delivery across distributed teams, time zones, and organisational layers.

Good to Have

 Prior experience with voice AI, ASR/TTS platforms, conversational AI, speech analytics, or NLP-based

products in production.

 Familiarity with LLM orchestration frameworks (LangChain, LlamaIndex, or similar), RAG pipelines, or multi-

agent systems.

 Cloud certifications (AWS Solutions Architect, GCP Professional Cloud Engineer, or equivalent).

 Experience with enterprise accounts in BPO, contact center, BFSI, healthcare, or government verticals in the US market.

 MLOps or LLMOps experience: model deployment pipelines, evaluation frameworks, and production AI

system operations.

 Familiarity with infrastructure-as-code tools (Terraform, Pulumi, or CloudFormation).