ML Engineer – Speech / Vision

Ovii's Interpretation of the Role

Deccan AI seeks a full‑time ML Engineer to design and optimize speech and vision models. You will build computer‑vision pipelines, speech‑to‑text and text‑to‑speech systems, and multimodal architectures that meet real‑world latency and accuracy constraints.

Role Snapshot

• Build computer‑vision models
• Develop speech processing pipelines
• Create multimodal AI systems
• Optimize models for latency & deployment
• Collaborate with frontier AI labs
• Work with Python & PyTorch/TensorFlow

Must-Have Requirements

• Python
• PyTorch/TensorFlow
• Computer Vision frameworks (OpenCV, YOLO, Detectron2)
• Speech processing frameworks (Whisper, Kaldi, ESPnet)
• Multimodal ML experience
• hands‑on experience in computer vision, speech processing, or multimodal ML

Nice-to-Have Signals

• Diffusion models, GANs, voice cloning
• experience with diffusion models, GANs, voice cloning

Work Setup

• Employment type: Full-Time

Not Specified in JD

• Remote eligibility
• Visa sponsorship
• Salary range
• Location
• Education requirement

What You'll Likely Work On

• Build and optimize computer‑vision models for detection, segmentation, and recognition
• Develop end‑to‑end speech pipelines covering ASR, TTS, and speaker diarization
• Design multimodal models that fuse vision, audio, and text representations
• Experiment with diffusion, GAN, and voice‑cloning approaches for vision/audio generation
• Tune models for real‑world latency, accuracy, and deployment constraints
• Partner directly with frontier AI labs to solve cutting‑edge perception problems

Good Fit If You Have

• Familiarity with diffusion, GAN, or voice‑cloning techniques
• Interest in high‑ownership, research‑adjacent engineering work
• Experience in AI infrastructure or large‑scale data pipelines