Forward Deployed Engineer

About the Role

We are building a local Enterprise Sales Pod in Bangalore with one Account Executive (AE) and one Forward Deployed Engineer (FDE). You will embed directly with enterprise customers to architect, build POCs, and ship production systems on Telnyx's global network for voice, messaging, AI, and wireless.

Your Responsibilities

• Embed with enterprise customers to understand workflows, AI use cases, and integration challenges firsthand across English and Indic languages (Hindi, Kannada, Tamil, etc.).
• Build and deploy custom implementations including AI Voice Assistants, Telnyx APIs (Voice, Messaging, Fax, Wireless), WebRTC, and Open-weight LLMs via Telnyx Inference.
• Deploy and operate LiteLLM as a model gateway in customer environments with routing, load balancing, retries, rate limits, and virtual keys for self-hosted models like BharatGPT or hosted providers.
• Govern LLM usage through cost tracking, caching, logging (OpenTelemetry), guardrails, and observability dashboards to ensure customers control their AI workloads.
• Wire production observability using Prometheus + Grafana, ELK/Graylog stacks for metrics, logs, traces, and alerting so customer ops teams understand system health.
• Design model routing strategies balancing latency, cost, and quality with sane fallbacks when providers degrade.
• Evaluate build-vs-buy trade-offs between self-hosted open-weight models and hosted frontier APIs to keep application code portable.
• Adapt models for customer domains using prompt engineering, RAG pipelines, evaluation harnesses, and compliance with Indian regulations like the DPDP Act and RBI cloud rules.
• Lead POCs from whiteboard to go-live, own outcomes until stability is achieved, and collaborate with Product/Engineering on roadmap insights.
• Create technical documentation and maintainable solutions for handoff while troubleshooting complex integration issues alongside customer teams.

We Are Looking For

• CS degree or equivalent experience.
• 3+ years building production software you've been on-call for, including debugging in real-world scenarios (consulting background counts).
• Proficiency in Python, Node.js/TypeScript, and Go.
• Practical understanding of model failure modes: provider rate limits, timeouts, token accounting, streaming behavior, and concurrency issues like consumer lag and backpressure.
• Comfort deploying containerized services on Kubernetes with secrets management, config, and upgrades.
• Experience wiring observability stacks (Prometheus, OpenTelemetry, ELK) to monitor health and alert effectively.
• High-concurrency experience with Kafka or message queues at scale.
• Built APIs from scratch including auth, rate limiting, versioning, and idempotency.
• Event-driven thinking and cloud-native instincts.
• Exposure to SIP, WebRTC, or real-time voice/messaging systems.
• Self-sufficient engineering judgment with no direct team behind you; ability to read docs/code and consult customers directly.
• Customer-facing experience running discovery sessions for engineers and presenting to executives simultaneously.
• Ability to translate "it's not working" into root cause analysis quickly.
• Comfortable on customer sites in high-stakes technical conversations.
• Excellent written/verbal English communication skills for whiteboarding, troubleshooting, and escalations.
• Based in Bangalore or willing to relocate for a hybrid role with travel across India.

Bonus Points

• Experience with AI voice assistants, STT/TTS, LLM-based conversational systems.
• Evaluating domains by building specific eval sets.
• Hands-on production experience with model gateways like LiteLLM or Portkey including config-driven definitions and virtual key management.
• Familiarity with open-weight models (BharatGPT, OpenHathi) and Indian sovereign AI initiatives.
• Inference serving engines behind the gateway: vLLM, SGLang, TGI for production; Ollama for on-prem.