Ixigo
Research Engineer - Agent Intelligence & Evaluation
New Delhi, DL, India · On-site · Full-time
Apply on the employer’s site ↗See how well you match this job
You apply directly with the employer. ROLIVA does not submit applications from this page and is not the employer.
About the role
Description from the employer’s posting. Check the employer’s page for the current version.
- EmployerIxigo
- LocationNew Delhi, DL, India
- Work styleOn-site
- Employment typeFull-time
- Employer posted
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.
Show full description
• Self-improvement systems. Mine production traces for failure patterns, generate targeted training or prompt data, validate fixes with adversarial replay, and guardrail against regressions.
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.
• You've shipped something non-trivial where research met production. You read papers, spot when a benchmark measures the wrong thing, and translate ideas from Interspeech, ACL, or NeurIPS into systems that run on real traffic. Publications welcome, not required.
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.
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.
Check your resume against this job Free, no sign-up. Nothing you paste is stored.
Staying safe while you apply
- Never pay to apply, interview or accept an offer. ROLIVA never asks you for payment to apply to a job.
- Apply only through the official “Apply on the employer’s site” button, on the employer’s own site, and check that a recruiter’s email uses that employer’s own domain rather than a free email address.
- Be cautious of an offer made with no real interview, or pressure to decide quickly or share financial details early.
Does something look wrong with this listing? Use “Report broken link” in the role details, or read ROLIVA’s full job-safety guidance ↗.
Preparing for this role with ROLIVA
- Check the fit.
Compare the requirements with the experience you have confirmed, including gaps.
- Prepare honestly.
Build interview stories and application drafts from your own evidence.
- Keep track.
Save the role and record your own application status and follow-ups.
At a glance
- Work styleOn-site
- Employment typeFull-time
- SeniorityMid level
- Employer posted
- Added to ROLIVA
- Last checked
Skills in this posting
Skills ROLIVA found in the employer’s own description above. A mention is not a requirement.