About the Role:

As a Founding AI Engineer, you will be one of the earliest technical hires and will have end-to-end ownership of the AI stack. This is not an execution-only role you will contribute to research direction, architecture decisions, and product-shaping conversations from day one. You will work directly with the founders and ship production-grade AI systems that are deployed at scale on global platforms.

If you are the kind of engineer who gets excited about reading papers on a Friday evening and has opinions on model architecture, this role is for you.

What You'll Own:

- Design, build, and maintain end-to-end multimodal AI pipelines covering image, video, audio, and text modalities

- Develop and fine-tune deep learning models for synthetic media detection, deepfake detection, and content classification

- Research and implement state-of-the-art techniques from academic literature and adapt them for production use cases

- Build scalable model training, evaluation, and inference infrastructure on cloud platforms (AWS / GCP / Azure)

- Collaborate with platform clients to understand labelling and moderation requirements and translate them into model objectives

- Establish AI engineering best practices: experiment tracking, model versioning, CI/CD for ML, and monitoring in production

- Mentor and grow the AI team as the company scales you will help define how we hire and onboard future engineers

- Contribute to technical blog posts, research notes, and external publications where relevant

Tech Stack & Tools:

- Frameworks: PyTorch, TensorFlow, HuggingFace Transformers, timm

- Multimodal models: CLIP, BLIP, Whisper, Stable Diffusion, ViT, DINO, LLaVA

- MLOps: MLflow, Weights & Biases, DVC, Docker, Kubernetes

- Cloud: AWS SageMaker / GCP Vertex AI / Azure ML

- Languages: Python (primary), with comfort in Bash and SQL

- Data pipelines: Apache Spark, Ray, or equivalent distributed processing frameworks

What We're Looking For:

Experience: 5+ years in AI/ML engineering with hands-on model development

- Strong foundations in deep learning CNNs, Transformers, attention mechanisms, contrastive learning

- Proven experience building and deploying multimodal models or computer vision systems in production

- Hands-on experience with model fine-tuning, distillation, quantisation, or other efficiency techniques

- Comfortable taking a research paper and turning it into working, production-ready code

- Experience with large-scale data labelling pipelines or content moderation systems is a strong plus

- Prior work on synthetic media, GAN-based detection, or adversarial robustness is highly desirable

- Excellent communication skills you can explain complex model decisions to non-technical stakeholders