Senior AI Engineer - Agentic AI & Voice Agents

YAL; Hyderabad; Sur place; INR 3 000 000 - 6 000 000; Plein temps

Résumé du poste

YAL in Hyderabad is seeking a senior AI engineer to design and lead agentic AI systems. The role focuses on building LLM agents that plan, call tools, and complete end-to-end business tasks across voice, chat and back-office channels.

You will stay hands-on while setting technical direction, owning the evaluation system to prove each agent’s readiness, and guiding a team of AI engineers to move use cases from demos to reliable, observable services clients trust with their customers.

Qualifications

• Experience designing and building LLM-driven agents for multi‑step business tasks.
• Knowledge of state/memory management, retrieval (RAG) and model routing.
• Ability to define tool schemas, permissions and human-in-the-loop checks.
• Experience creating robust workflows with checkpoints, retries and idempotency.

Responsabilités

• Design and build LLM agents that plan, call tools and complete multi-step tasks across voice, chat and back-office channels.
• Implement task and conversation state, memory, context management and retrieval (RAG) with model routing.
• Define tool schemas and permissions; validate tool calls and add human approval points for high-impact actions.
• Build durable workflows with checkpoints, retries, idempotency keys, timeouts and compensation for failures.
• Use multi-agent designs only where they beat a single agent.
• Integrate CRMs, ticketing systems and databases with warm human handoff.

Connaissances

LLM agents

Multi-agent systems

Tool orchestration

Speech processing

Outils

Twilio

Exotel

WebRTC

ASR

TTS

Description du poste

We are hiring a senior AI engineer to build and lead agentic AI systems: LLM agents that plan, call tools

and complete real business tasks end to end. You will build agents that hold live phone and web

conversations — streaming speech recognition, reasoning and tool calls, and speech synthesis in real

time. You will stay hands‑on while setting technical direction, owning the evaluation system that proves

each agent works before every release, and helping a team of AI engineers move use cases from a demo

to a reliable, observable service that clients trust with their customers.

Key responsibilities

Agentic systems

• Design and build LLM agents that plan, call tools and complete multi‑step business tasks across voice, chat and back‑office channels.
• Implement task and conversation state, memory and context management, retrieval (RAG) and model routing between fast and more capable models.
• Define tool schemas and permissions, validate every tool call, and add human approval points for high‑impact actions.
• Build durable workflows with checkpoints, retries, idempotency keys, timeouts and compensation for failed actions.
• Use multi‑agent designs, such as a router with specialist agents, only where they measurably beat a single agent.
• Integrate CRMs, ticketing systems and databases for actions such as complaint registration, classification, routing, status updates and notifications, with warm human handoff.

Voice agents

• Build streaming pipelines that connect telephony (SIP, Twilio, Exotel) and WebRTC audio to ASR, LLMs and TTS over WebSockets.
• Implement voice activity detection, turn detection, barge‑in and cancellation so the agent stops talking the moment a caller speaks.
• Set and hold latency budgets per hop (ASR, LLM, TTS, network), with a target of under [1 s] from end of caller speech to first agent audio.
• Adapt agent behaviour for speech: short turns, confirmations before actions, and recovery from mishearing.
• Support multilingual and code‑switched calls, including language detection and matching TTS.

Evaluation

• Build and maintain golden test sets that define the user goal, persona, expected tool calls and pass criteria for each scenario.
• Run trajectory evals that check each step, tool choice, arguments and order, not just the final answer.
• Run simulated‑caller evals for voice: an LLM plays the caller, TTS voices it with accents, noise and interruptions, and the agent is tested end to end over real audio.
• Score task success, tool‑call correctness, policy compliance, hallucination, word error