Research Engineer - Agent Intelligence & Evaluation

ixigo; Delhi; On-site; INR 1,400,000 - 2,000,000; Full time

Benefits offered by this job

Meaningful equity

Autonomy over tooling

Support to publish

Job summary

ixigo is building self-healing voice agents for enterprise customer support. This fellowship owns the intelligence layer: the evals that catch failures before shipping, the observability that traces them across the pipeline, and the feedback loops that let agents fix themselves.

You'll work with senior ML engineers to build robust evaluation metrics, real-time telemetry, and automated data pipelines, with equity and autonomy over tooling as part of a small, high-impact team.

Qualifications

• 3 to 5 years in ML engineering, research engineering, or applied research.
• Strong Python and modern ML tooling.
• Depth in at least two of: speech and audio models, LLM agent systems, and eval or observability infrastructure.

Responsibilities

• Evaluation infrastructure: audio-native metrics for barge-in, prosody, and turn-taking; adversarial datasets across accents and edge cases; LLM-as-judge rubrics for task success, tool-use correctness, and recovery.
• Observability across the pipeline: tracing that correlates audio, STT, LLM reasoning, tool calls, and TTS to a single conversation; analysis and alerting that surfaces cascade failures.
• Self-improvement systems: mine production traces for failure patterns, generate targeted training or prompt data, validate fixes with adversarial replay, and guardrail against regressions.

Skills

Python

ML tooling

Speech models

LLM agent systems

Eval infra

Tools

PyTorch

TensorFlow

Airflow

Job description

• Full-time

Company Description

We’re building self-healing voice agents for enterprise customer support within ixigo. The system has to know when it’s failing, why it’s failing, and how to fix itself before a human notices. This fellowship sits at the intelligence layer behind that work.

Job Description

Voice agents fail in ways traditional software doesn't. ASR confidence drops on an accent and a tool call misfires. Latency breaks turn-taking and the LLM hallucinates a policy. A model swap silently regresses production and nobody catches it for a week.

We're building self-healing voice agents for enterprise customer support. This role owns the intelligence layer: the evals that catch failures before shipping, the observability that traces them across the pipeline, and the feedback loops that let agents fix themselves

What you'll own

• Evaluation infrastructure. Audio-native metrics for barge-in, prosody, and turn-taking. Adversarial datasets across accents and edge cases. LLM-as-judge rubrics for task success, tool-use correctness, and recovery.
• Observability across the pipeline. Tracing that correlates audio, STT, LLM reasoning, tool calls, and TTS to a single conversation. Analysis and alerting that surfaces cascade failures instead of hiding them.
• Self-improvement systems. Mine production traces for failure patterns, generate targeted training or prompt data, validate fixes with adversarial replay, and guardrail against regressions.

Qualifications

Who we're looking for 3 to 5 years in ML engineering, research engineering, or applied research. Strong Python and modern ML tooling. Depth in at least two of: speech and audio models, LLM agent systems, and eval or observability infrastructure.

Nice to have Real-time systems or telephony experience. Work on RLHF, DPO, or synthetic data pipelines. Familiarity with enterprise deployment (SOC 2, PII, data residency).

What you'll get Senior seat on a small team where research and production aren't separate orgs. Real enterprise conversation data under proper governance. Meaningful equity, autonomy over tooling, and support to publish.

Additional Information

Candidates are responsible for safeguarding sensitive company data against unauthorized access, use, or disclosure, and for reporting any suspected security incidents in line with the organization's ISMS (Information Security Management System) policies and procedures.