Ai Ml Engineer

Role & responsibilities :

• RAG Pipeline Development: Design, build, and maintain production-grade Retrieval-Augmented Generation (RAG) pipelines.
• Data Preparation & Vectorization: Implement advanced chunking strategies, select optimal text embedding models, and manage vector databases for highly accurate retrieval.
• Search & Retrieval Optimization: Fine-tune the retrieval process using advanced search, hybrid search, and reranking models to ensure the most relevant context is fed to the LLM.
• Core Machine Learning: Apply classic ML techniques where appropriate, including model selection, regression, classification, clustering, and ensemble methods (bagging and boosting).
• System Evaluation & Monitoring: Build internal leaderboards and evaluation frameworks to rigorously track and capture pipeline performance, response accuracy, and token costs.
• Integration & Deployment: Consume and integrate LLMs via robust REST APIs or Model Context Protocol (MCP) based integrations to ensure seamless data flow across applications.

Preferred candidate profile :

Generative AI & RAG Orchestration

• Chunking & Embeddings: Deep understanding of semantic, fixed-size, and parent-child chunking strategies alongside modern text embedding models.
• Search Architecture: Hands-on experience with vectorization, vector databases (e.g., Pinecone, Milvus, Qdrant, Chroma), and deploying retriever and reranking algorithms (e.g., Cohere Rerank, Cross-Encoders).
• LLM Utilization & Optimization: Proven track record of consuming LLMs (OpenAI, Anthropic, open-source via Hugging Face) while implementing strict token optimization techniques to control latency and costs.
• Integration Protocols: Experience with API development and integration, including standard RESTful architectures and emerging standards like MCP (Model Context Protocol).

Core Machine Learning

• Solid grasp of foundational ML concepts: Model selection, regression, classification, and clustering.
• Strong hands-on experience with ensemble methods, specifically bagging and boosting (e.g., Random Forest, XGBoost, LightGBM).

Analytics & MLOps

• Experience building analytics dashboards or leaderboards to monitor system performance, output accuracy (e.g., RAGAS framework, TruLens), and financial token metrics.