AI/ML Engineer

About the Company

Spekond Technology is building a child developmental assessment platform, powered by advanced AI. We are looking for experienced engineers to help us scale our core AI pipeline and deliver intelligent solutions.

Responsibilities

• Design and build end-to-end AI pipelines for multimodal data processing including video, audio, image, and text.
• Work on recommendation models, neural networks, and LLMs.
• Integrate and work with Large Language Models and Vision Language Models for intelligent data extraction and reasoning.
• Build and maintain RAG pipelines for grounded, context-aware AI outputs.
• Develop and orchestrate agentic AI workflows for complex multi-step automation.
• Build recommendation systems and personalization engines based on user behavior and segmentation.
• Deploy and maintain AI models in production cloud environments with proper monitoring and evaluation.
• Expose AI capabilities as scalable REST APIs for product integration.
• Perform exploratory data analysis and derive insights to support product decisions.
• Collaborate with cross-functional teams to translate requirements into AI solutions.
• Build evaluation and testing frameworks to ensure model quality, reliability, and safety.
• Stay current with the latest advancements in LLMs, multimodal AI, and agentic systems.

Requirements

• Generative AI and LLMs: Hands-on experience building RAG pipelines (document processing, chunking strategies, embedding models, vector databases), experience with Vision Language Models, strong prompt engineering skills (system design, few-shot learning, chain-of-thought, structured output), experience with LangChain and LangGraph or similar orchestration frameworks, familiarity with LLM API integration and multi-model pipeline design, understanding of hallucination mitigation and output grounding strategies.
• Machine Learning and Deep Learning: Strong foundation in NLP (text classification, sequence modeling, transformer-based architectures), experience fine-tuning or working with pretrained language models, hands-on experience with graph-based learning or relational modeling, experience with multi-label and multi-class classification problems, proficiency in PyTorch and/or TensorFlow and Keras, experience with MongoDB.
• MLOps and Deployment: Docker for containerization, GCP Cloud Run or equivalent serverless deployment, CI/CD pipeline setup with GitHub Actions or similar, FastAPI for exposing ML model APIs, model versioning, monitoring, and production maintenance.
• Soft Skills: Strong problem-solving mindset, ability to work on ambiguous problems, experience taking AI systems from research to production, self-driven, ownership mindset, strong communication skills.