About this role
Engineers who thrive in autonomous, client-facing environments and enjoy turning complex connectivity and AI concepts into production systems will find a strong fit here. Telnyx is expanding its enterprise presence in the United Arab Emirates by building a local pod model in Dubai, pairing an enterprise account executive with a technical partner. This role centers on embedding directly with enterprise clients to architect, build, and ship scalable solutions spanning voice, messaging, wireless, and artificial intelligence.
The person in this position will handle the entire technical journey from initial discovery and custom proof-of-concept design to production go-live. Responsibilities include running open-weight large language models, deploying model gateways, wiring comprehensive observability stacks, and addressing regional data sovereignty requirements. Success requires a self-sufficient builder who can seamlessly switch between client discovery sessions, executive presentations, and hands-on root-cause debugging without relying on a behind-the-scenes engineering support layer.
What you will do
- Embed directly with enterprise customers to understand communications workflows, artificial intelligence use cases, and integration challenges firsthand.
- Build and deploy custom implementations including AI voice assistants, Telnyx application programming interfaces, and WebRTC protocols.
- Run open-weight large language models in client environments or deploy self-hosted stacks when air-gapped requirements demand it.
- Instrument and govern language model usage through gateways to manage cost tracking, caching, logging, observability, and safety guardrails.
- Wire production observability for all shipped workloads using metrics, logs, traces, dashboards, and alerting tools.
- Lead proof-of-concept initiatives, pilots, and production launches from the initial whiteboard phase to stable go-live operations.
- Collaborate directly with product and engineering teams to shape roadmaps based on field insights gathered across the GCC region.
What you need
- Hold a computer science degree or possess equivalent practical software engineering experience.
- Bring at least 3 years of experience building and shipping production software that real users depend on, including on-call rotation and debugging.
- Demonstrate proficiency in multiple programming languages such as Python, Node.js, TypeScript, or Go.
- Possess a practical understanding of production failure modes in front of models, including rate limits, timeouts, streaming, and concurrency issues.
- Feel comfortable deploying containerized services on Kubernetes with proper secrets management, configuration, and upgrades.
- Demonstrate high-concurrency experience handling message queues or event streams at scale to manage consumer lag and backpressure.
- Communicate excellently in both English and Arabic to run whiteboard sessions, live troubleshooting, and executive presentations.
- Be based in Dubai or willing to relocate for this hybrid role that includes travel across the UAE and broader GCC.
Nice to have
- Experience with AI voice assistants, speech-to-text, text-to-speech, or LLM-based conversational systems
- Building evaluation sets for specific domains
- Hands-on production experience with a large language model gateway such as LiteLLM, Portkey, or Kong AI Gateway
- Familiarity with regional regulatory environments including UAE PDPL and Dubai data sovereignty
Skills and tech stack
Questions candidates ask
What is the work arrangement for this role?
This is a hybrid role based in Dubai, or open to those willing to relocate, with required travel across the UAE and broader GCC.
What kind of team structure will I be working within?
You will work as part of a local Enterprise Sales Pod in Dubai, partnering directly with an enterprise account executive rather than operating within a large traditional engineering hierarchy.
What technical stack will I use for observability and model routing?
You will deploy tools like Prometheus and Grafana for metrics, OpenTelemetry for traces, Graylog or ELK for logs, and LiteLLM as the model gateway in customer environments.
Will I have a dedicated engineering team behind me for support?
You are self-sufficient by default with no engineering team behind you, meaning you read the code and documentation, ask the customer, and make sound engineering judgment calls on your own.