한국어·English
Telnyx company · OPEN ROLE
- Company
- Own the network, models and agent code; turn enterprise conversations into working systems.
- Role
- Own discovery, architecture, gated POCs and stable production beside an enterprise AE.
Log in to mock apply, save this role or see more roles.
What you will do
We rewrote the responsibilities and qualifications in plain language instead of copying the original posting.
Responsibilities
- Embed with enterprise customers, discover communications/AI/integration needs and own architecture, gated POCs and production go-live.
- Work with a dedicated enterprise AE: the AE owns accounts, executive relationships and close; the FDE owns technical discovery, architecture, the first live workload and finding the next.
- Build and deploy AI voice assistants, Voice/Messaging/Fax/Wireless APIs and WebRTC in customer environments.
- Use GLM, Kimi, DeepSeek, Qwen and MiniMax through Telnyx Inference, or self-host Llama, Mistral and Korean open-weight models when air-gapped or sovereign requirements demand it.
- Deploy and operate LiteLLM with routing, load balancing, retries/fallbacks, rate limits and per-team virtual keys.
- Instrument and govern LLM usage with cost tracking, budgets, caching, logging, observability and guardrails.
- Wire production metrics, logs, traces, dashboards and alerting so the customer operations team can recognize health and diagnose failures.
- Design real-time voice model routing to balance latency, cost and quality with sensible fallback when providers degrade.
- Explain build-versus-buy choices between self-hosted open-weight models and hosted frontier APIs while keeping application code portable.
- Adapt prompts, RAG and evaluation to Korean/English and Korea/APAC customer environments; the employer names PIPA, data-localization and PSC cloud-security requirements.
- Stay engaged until customer systems are live and stable and share South Korea/SEA field insights with Product and Engineering.
- Create documentation, runbooks and maintainable handoffs and troubleshoot complex integrations with customer teams.
What you need
- CS degree or equivalent experience is required.
- 3+ years building and shipping production software used by real users, including on-call ownership and debugging; consulting counts when those conditions hold.
- Proficiency in multiple programming languages; the posting names Python, Node.js/TypeScript and Go, with TypeScript also used for Edge Compute.
- Practical understanding of provider rate limits/quotas, timeouts/retries, streaming, token accounting/cost attribution and failures under concurrency.
- Comfort deploying containerized services on Kubernetes with secrets, configuration and upgrades.
- Have wired observability and been paged using Prometheus/Grafana, OpenTelemetry, Graylog/ELK or equivalent tooling.
- Real-scale high-concurrency experience with Kafka, queues or event streams, including lag, backpressure, hot partitions, rebalance behavior, partitioning and consumer groups.
- Experience building APIs from scratch, including OpenAPI, REST/GraphQL, authentication, rate limits, versioning and idempotency.
- Event-driven, cloud-native thinking and exposure to SIP, WebRTC or real-time voice/messaging systems.
- Be self-sufficient, make sound engineering decisions, run discovery with customer engineers, present to executives, work on customer sites and find root causes.
- Excellent written and verbal English and Korean, including whiteboard sessions, live troubleshooting and escalations with customer engineering teams.
- Based in Seoul or willing to relocate; a hybrid role with travel across South Korea and the broader SEA region.
- Legally authorized to work in South Korea or eligible for sponsorship.
Good to have
- AI voice assistants, STT/TTS or LLM conversational systems and building domain-specific evaluation sets are bonuses.
- Production LLM gateway experience with LiteLLM, Portkey, Kong AI Gateway or an in-house OpenAI-compatible proxy, including model definitions, routing/fallback and virtual keys, is a bonus.
- Familiarity with regional/multilingual open-weight tradeoffs, EEVE-Korean/Polyglot-KR/Kanana and Korean sovereign AI efforts such as Naver, KT and SK Telecom is a bonus.
- Serving-engine experience with vLLM, SGLang or TGI for production, Ollama for lightweight on-prem, and GPU sizing for KV cache/batches/context versus VRAM is a bonus.
- SQL proficiency with Postgres/MySQL/Oracle and ETL/data-wrangling experience are bonuses.
- CI/CD design/automation and telecom, CPaaS or hypergrowth SaaS backgrounds are bonuses.
- Sovereign/on-prem deployments and Korea/APAC regulatory experience are bonuses; the employer names PIPA, PSC, location-data localization, FSC and cross-border frameworks.
- A security mindset covering IAM, encryption and audit logging is a bonus.
Read the employer’s original job description
See how your resume reads for this role
A former Korea GM at a global company reads it and tells you what to improve for this role. It uses 1 credit.
Checked 2026-10-17
TelnyxForward Deployed Engineer, SEAStart a mock application