Freelance ML Engineer | Speech Model Training, Quantization & Deployment

Job Description

geteniac.tech is looking for a

freelance Machine Learning Engineer

to help train, optimize, and deploy

speech models

for client projects.
You’ll work on preparing and serving customized

speech-to-text

and

text-to-speech

models, ensuring low latency, high reliability, and seamless integration into production environments.

Responsibilities

Train and fine-tune open-source

speech models

such as Whisper, Wav2Vec2, or MMS
Quantize models

(INT8, FP16, GGUF, TensorRT, etc.) for optimized inference performance
Deploy models

with low latency using scalable serving infrastructure
Manage

ML lifecycle

including experiment tracking and model versioning
Utilize

vLLM, Ray, and MLflow

to ensure robust and efficient deployment pipelines

Requirements

Proven experience in

training and deploying speech or LLM-based models
Strong understanding of quantization and inference optimization
Hands-on with vLLM, Ray, MLflow, Docker, and Kubernetes
Familiarity with FastAPI

and

Triton Inference Server

is a plus
Ability to independently manage

end-to-end ML pipelines

Commitment

10–12 hours per week
Remote and flexible schedule
Flexible hourly pricing

based on experience and project scope

About Geteniac

geteniac.tech helps businesses

automate phone calls with AI .

From instant phone number generation to natural, AI-driven conversations, our platform enables companies to manage inbound and outbound calls efficiently and at scale.

We’re now expanding our capabilities with

custom speech model deployments

to meet specific client requirements.

How to Apply

If you have relevant experience and want to work on real-world speech model projects, reach out at

ranga.sumit1999@gmail.com

or message me directly here on LinkedIn.