Brahma

• Research and develop state-of-the-art voice synthesis models (e.g., TTS, voice cloning, speech-to-speech). • Build and fine-tune models using frameworks like PyTorch and HuggingFace. • Design training pipelines and datasets for scalable voice model training. • Explore techniques for emotional expressiveness, multilingual synthesis, and speaker adaptation. • Work closely with product and creative teams to ensure models meet quality and production constraints. • Stay on top of academic and industrial trends in speech synthesis and related fields.

• Own the unified design language for the BRAHMA AI platform. • Align designers across the different product teams. • Lead the design of our generative AI tooling. • Define end-to-end user journeys across the platform.

• Join our team and help build state-of-the-art generative systems for video and audio synthesis, performance transfer, and visual translation. • Shape core infrastructure and set best practices. • Communicate with stakeholders and mentor junior engineers. • Deliver high-quality ML pipelines in production environments.

• Lead end-to-end design for AI-powered creative tools and workflows, from discovery through implementation • Own and maintain all design prototypes, flows, and specifications in Figma, serving as the source of truth for product and engineering teams • Define and evolve the user experience for complex generative AI features including digital humans (ATMAN), voice synthesis (VAANI), and content management systems • Prototype and ship functional React components quickly, using AI-assisted coding tools • Build and maintain our React-based design system and component library, ensuring consistency across products • Create interactive prototypes using code (React, TypeScript) to validate ideas and communicate design intent precisely • Implement production-ready UI components, working alongside frontend engineers • Conduct user research and usability testing to validate design decisions and uncover opportunities • Champion design quality in code reviews and contribute to frontend architecture decisions • Present design concepts and rationale to stakeholders, incorporating feedback effectively

• Feature engineering at scale • Architect unstructured data pipelines • Orchestrate ML workflows using Dagster or Airflow • Optimize compute & cost • Build "dataset-as-code" • Infrastructure ownership with Kubernetes for scalability

• Own the end to end audio workflow, from raw recordings to final mastered outputs • Perform professional editing, cleanup, restoration, mix, and mastering across dialogue, voice, and sound assets • Curate, clean, annotate, and structure high quality audio datasets for AI voice model training • Train, operate, and iterate on AI voice models across text to speech and voice to voice systems • Evaluate model outputs critically and refine datasets, prompts, and processing to improve voice quality • Act as the audio quality gatekeeper, ensuring consistency, clarity, and realism across all deliverables • Collaborate closely with AI engineers, developers, and creative teams to align technical outputs with creative intent • Define and document audio standards, workflows, and best practices • Troubleshoot complex audio and pipeline issues across different tools and formats • Stay current with emerging audio, voice AI, and production technologies and contribute to R&D efforts

• Design and implement scalable backend architectures for AI-powered creative tools and workflows. • Build high-performance APIs and microservices that handle generative AI inference/training and video/audio processing. • Design and implement user interfaces that make complex generative AI capabilities intuitive and accessible. • Develop robust data pipelines for ingesting, processing, and storing large-scale media content. • Design and maintain database schemas, caching layers, and data storage solutions for multimedia content. • Build authentication, authorisation, and security systems for enterprise-grade applications. • Optimise backend performance for low-latency AI inference and high-throughput media processing. • Integrate backend systems with ML training and inference pipelines. • Implement monitoring, logging, and alerting systems for production backend services. • Collaborate with ML engineers to optimise model serving and deployment infrastructure.

• Own build, deploy, and runtime reliability across BRAHMA AI’s hybrid estate. • Deliver secure, scalable infrastructure for Gen AI based workflows and products across hybrid environments. • Partner with infrastructure and multidisciplinary product and research teams to help them innovate and ship fast. • Design, implement, and operate Slurm and Kubernetes-based platforms across cloud and on-prem GPU nodes, including autoscaling, rollout strategies, and multi-cluster operations. • Build CI/CD pipelines for services, model training, and model serving; standardise artifact/version management and environment promotion. • Implement Infrastructure as Code with Terraform/Terragrunt and configuration management; enforce drift detection and repeatable environments. • Design and implement observability stacks (metrics, logs, tracing); drive incident response and postmortems. • Secure the stack with least privilege, secrets management, network policy, and hardened baselines; support ISO/MPA controls with the security team. • Operate model-serving infrastructure for real-time and batch workloads; optimise GPU utilisation, concurrency, and latency. • Drive cost visibility and efficiency across compute, storage, and egress; forecast capacity and plan lifecycle of hardware and licenses.

• Research and develop cutting-edge generative video models, with a focus on controllable facial expression, head motion, and audio-driven lip synchronisation. • Fine-tune and extend video diffusion models such as WAN and Hunyuan for better visual realism and audio-visual alignment. • Design robust training pipelines and large-scale video/audio datasets tailored for talking-head synthesis. • Explore techniques for controllable expression editing, multi-view consistency, and high-fidelity lip sync from speech or text prompts. • Work closely with product and creative teams to ensure models meet quality and production constraints. • Stay current with the latest research in video generation, speech-driven animation, and 3D-aware neural rendering.

• Design and maintain efficient, functional workflows for deploying audiovisual GenAI models, primarily using ComfyUI. • Identify, integrate, and update workflows with state-of-the-art tools and technologies. • Collaborate with GenAI and research teams to develop, evaluate, and refine workflow templates that maximize model performance. • Experiment with novel model and tool combinations to enhance existing workflows and create new ones. • Support creative and product teams by building templatable, tailored workflows and working with multi disciplinary teams to transition them into scalable solutions. • Present and evaluate alternative model deployment platforms while contributing to internal platform development. • Onboard and train artists from other disciplines in ComfyUI workflow usage.