Artificial Intelligence Engineer

About Trinetra AI

Trinetra AI is pioneering a state-of-the-art Generative AI platform tailored for filmmakers, content creators, studios, marketers, and enterprises. Our mission spans the entire content creation process—from ideation, scripting, and storyboarding to character design, image/video generation, voice dubbing, music production, editing, and final output. Positioned at the crossroads of Generative AI, multimodal intelligence, video technology, and SaaS, we focus on translating cutting-edge AI research into impactful, scalable products.

Role Overview

We seek a proactive AI Engineer adept at designing, developing, integrating, fine-tuning, and optimizing Generative AI systems. This role requires deep expertise beyond traditional machine learning, capable of advancing AI capabilities from research and prototypes to scalable production deployments. Responsibilities span LLMs, VideoGen, ImageGen, Audio/Voice AI, multimodal models, AI agents, and GPU acceleration.

Key Responsibilities

• Create AI functionalities across text, image, video, voice, audio, and music domains.
• Conduct experiments with foundational models and emerging AI structures.
• Assess models based on quality, latency, cost-efficiency, licensing, and scalability.
• Develop features involving text-to-video/image, voice generation, dubbing, lip-sync, and maintaining character consistency.
• Fine-tune and customize foundational models using proprietary datasets.
• Design and implement production-grade large language model (LLM) applications, including prompt engineering, retrieval-augmented generation, embeddings, tool integration, context handling, model routing, fine-tuning (LoRA/QLoRA), and guardrail development.
• Handle complex video and multimodal AI tasks using diffusion models, vision transformers, video generation techniques, and ensure temporal and character consistency with advanced conditioning controls.
• Establish pipelines for dataset preparation, augmentation, fine-tuning, distributed training, checkpointing, hyperparameter tuning, and evaluation.
• Collaborate cross-functionally with engineering, product, data, and content teams to transform large media datasets into quality AI training inputs.
• Develop evaluation frameworks for visual quality, consistency, lip-sync accuracy, model accuracy, latency optimization, and cost reduction.
• Optimize models with quantization, batching, model compilation, caching, mixed precision, parallel inference, and GPU memory management.
• Work extensively with GPU infrastructure including NVIDIA A100/H100/H200, multi-GPU setups, distributed training, Kubernetes GPU workloads, training clusters, and inference environments.
• Architect scalable AI microservices, APIs, asynchronous inference pipelines, job queues, orchestration services, and GPU scheduling mechanisms.
• Create AI workflow orchestration systems integrating multiple AI models and agents to support comprehensive content creation pipelines from idea to final output.
• Lead efforts in reading research papers, reproducing experiments, benchmarking, modifying pipelines, and deploying effective AI prototypes into production.

Required Technical Skills

• Advanced proficiency in Python programming.
• Experience with PyTorch and Transformer architectures.
• Working knowledge of the Hugging Face ecosystem.
• Strong foundation in Generative AI and Large Language Models.
• Expertise in deep learning methodologies.
• Building REST APIs using FastAPI.
• Version control with Git and Linux environment usage.
• Containerization with Docker.
• GPU-based model inference and training.
• Techniques for fine-tuning models, including LoRA and QLoRA.
• Model quantization and prompt engineering.
• Comprehensive model evaluation and benchmarking.

Preferred Additional Experience

• Familiarity with Diffusers, ComfyUI, Stable Diffusion, FLUX, ControlNet, IP-Adapter.
• Computer vision tools such as OpenCV and FFmpeg.
• Expertise in video generation models.
• Speech technologies including Whisper, TTS, voice cloning.
• CUDA programming and TensorRT optimization.
• Knowledge of vLLM, Triton, DeepSpeed, ONNX frameworks.
• Experience with LangGraph, LangChain, LlamaIndex.
• Exposure to cloud AI infrastructure like AWS, GCP, or Azure.
• Familiarity with databases and orchestration tools such as PostgreSQL, MongoDB, Redis, Vector DBs, Docker, and Kubernetes.

Candidate Profile

The ideal candidate has 3-7 years of software, AI, or machine learning engineering background with significant hands-on experience in Generative AI. Previous involvement in building AI products for real users and working with open-source foundation models is essential. Candidates should be adept at training or fine-tuning models and deploying GPU-powered inference workloads. A solid engineering mindset, problem-solving ability, and comfort working in fast-moving startup environments are crucial.

What Success Entails

• Ability to independently assess and integrate new foundational AI models.
• Deliver production-ready AI APIs and solutions.
• Utilize proprietary datasets for model fine-tuning to enhance quality and consistency.
• Troubleshoot hallucinations and model inaccuracies effectively.
• Optimize inference latency, GPU utilization, and production costs.
• Define and maintain benchmarks that accurately capture model performance.
• Transform AI research advances into innovative product features and contribute to proprietary intellectual property.

Additional Information

Working with Trinetra AI offers a unique opportunity to reshape film, advertising, and digital content creation through AI innovation. The role involves tackling real-world challenges in multimodal AI, video generation, AI filmmaking, and large-scale GPU-powered production systems. Join us to contribute to a global-scale Generative AI platform crafted from India.