B

AI Data Engineer (Speech & Language)

Blue Machines AI
NewPosted today

LOCATION

Bengaluru · Onsite

EXPERIENCE

1 - 2 Years

SKILLS REQUIRED

Data PipelinesPythonData QualityData LineageSpeech RecognitionNatural Language Processing

Job description

Overview

Own the data that trains and evaluates every model the AI team builds, covering pretraining, supervised fine-tuning (SFT) and reinforcement learning (RL) runs for speech-to-text (STT), large language models (LLM), text-to-speech (TTS), speech-to-speech (S2S), language identification (LID) and intent prediction, all built for Indian languages and code-mixed speech and language.

What you'll do

  • Build training data for every stage of the model lifecycle: prepare corpora for continued pretraining, instruction and conversation data for SFT, and preference pairs, reward signals and rollout data for RL (DPO, GRPO and similar).
  • Own the data harvesting pipeline end to end, collecting from public web and speech sources, licensed datasets and consented production logs.
  • Clean, deduplicate, filter, redact PII and track provenance and licensing for every sample.
  • Generate synthetic data at scale: use LLMs to create code-mixed dialogues, intent-labelled utterances and domain conversations, and use TTS and audio augmentation (noise, telephony codecs, speed, reverb) to grow speech coverage.
  • Make data model-ready: output the exact format each training stack expects, including chat-templated JSONL, ASR/TTS manifests with aligned transcripts and durations, and sharded and packed datasets, with versioning and lineage so every run is reproducible.
  • Build the evaluation pipeline for each model: curate held-out and production-like test sets per language and domain, and automate metrics. STT: WER/CER per language, code-mixed and entity error rates. TTS: intelligibility via ASR round-trip, pronunciation checks, MOS collection. LID: accuracy and confusion across languages and scripts. Intent: precision, recall, F1 per class. LLM: task accuracy, LLM-as-judge rubrics, regression suites.
  • Close the loop with production: learn how the voice agent behaves in real calls, mine failure cases from logs, and turn them into new training and test data.
  • Handle Indic text properly: Unicode normalization (nukta, ZWJ/ZWNJ, matras), numerals and dates, inverse text normalization, transliteration between native script and romanized text, and consistent treatment of code-switched Hindi–English and other language pairs.
  • Ship fast and document: build small, tested, reusable pipeline components and clear dataset cards stating what went in, why, and in what proportion.

What you'll need

  • 1–2 years building data pipelines, ML pipelines or backend systems; strong internships or open-source work count.
  • Strong Python, plus comfort with Linux, Bash, Git and processing large text and audio files efficiently.
  • Working knowledge of how STT, TTS, LLM and classifier models are trained, and what good training data looks like for each (alignment, duration limits, speaker balance, label quality, chat templates).
  • Hands-on use of Hugging Face Datasets, pandas or Polars, and audio tooling such as librosa, torchaudio, sox or ffmpeg.
  • Experience running parallel or distributed jobs (multiprocessing, Ray, Spark or Dask) and working with cloud object storage.
  • Familiarity with Indian languages and scripts: can spot a broken Unicode sequence, a wrong transliteration or a mislabelled language.
  • Read or speak at least one Indian language besides English.
  • An engineering mindset: test your pipelines, measure data quality with numbers, and ship a working first version within days, not weeks.

Nice to have

  • Experience with Indic NLP tooling such as IndicNLP library, IndicTrans2, Aksharamukha or indic-transliteration, and with datasets like IndicVoices, Kathbath or Sangraha.
  • Forced alignment and segmentation (CTC alignment, Montreal Forced Aligner, VAD-based chunking) for long-form audio.
  • Ensemble or model-in-the-loop labelling: combining several ASR or LLM outputs to auto-label data and filter low-confidence samples.
  • Quality filtering techniques: MinHash deduplication, model-based quality classifiers, test-set contamination checks.
  • Exposure to training frameworks the data feeds into: NeMo, ESPnet, Hugging Face TRL, Unsloth or Axolotl.
  • Annotation tooling (Label Studio or similar) and running small human-labelling projects.

Details

  • Location: Bengaluru, Karnataka, India.

Read the full description and apply on the company’s own careers page.

Stay safe

Hiring on Abekus is free for applicants

We never charge a fee, and employers are prohibited from doing so. If a recruiter asks for payment, please report them right away.

AI Data Engineer (Speech & Language)