Senior AWS Engineer

At IdeaSouq, we are building the "AI operating system" to transform how private market investors and funds discover, evaluate, and manage their opportunities. Traditional investment workflows are drowning in data silos, manual screening, and overwhelming deal flow. We're a startup building the solution: an AI analyst that turns this deal flow chaos into funding intelligence. Our platform provides smart sourcing, deep evaluation based on a fund's unique thesis, agentic workflows to find missing data, and seamless portfolio monitoring, all while built on a foundation of enterprise-grade security and data sovereignty . We are looking for a foundational engineer to join our small, high-impact team. If you are passionate about building secure, scalable, and self-evolving AI systems from the ground up, this is your opportunity to define the future of investment technology. As our Senior LLM Engineer, you are the engine behind our "AI analyst." You will bridge the critical gap between abstract "state-of-the-art" LLM research and the concrete, production-grade data engineering required to make it work at scale. You will be responsible for building the entire data and model lifecycle that allows our system to ingest, understand, and reason over vast, unstructured datasets (pitch decks, financial reports, transcripts, web data). You are the kind of person who reads a new paper on agentic RAG on Monday, designs the data pipeline for it on Tuesday, builds the prototype by Thursday, and works with our cloud team to get it into production. This is not a pure research role; it is a hands-on "builder" role for someone who scans arXiv but builds for enterprise-scale, secure deployment. Data Engineering for AI Data Pipelines: Design, build, scale ingestion pipelines (unstructured/structured data). Vector DBs: Architect, manage, optimize vector stores (Milvus, Pinecone, AI Search). ETL/ELT for AI: Build cleansing workflows; turn "deal flow chaos" into model-ready data. Engineer data systems for continuous AI learning & versioning. Applied LLM Research & Implementation Applied Research: Implement SOTA research (LLMs, RAG, agentic frameworks). Fine-Tuning: Tune open-source models (Llama, Mistral) on proprietary & client data. Build data backbone for "Agentic Clarifications" (AI finds missing info). Model Evaluation: Develop frameworks to measure model quality, accuracy, cost, latency. We are looking for a senior-level candidate with a proven track record of building data-intensive AI products. Required Experience: ~5+ years of professional experience in a role blending Data Engineering and Machine Learning. ~ Proven, expert-level experience in building and managing large-scale, production-grade data engineering pipelines (e.g., using Spark, Airflow, Kafka, or cloud-native equivalents). ~ Deep, hands-on experience with the end-to-end LLM lifecycle: from data preprocessing and fine-tuning to production deployment. ~ Demonstrable ability to read, understand, and implement novel techniques from AI/ML research papers (arXiv, NeurIPS, etc.). Programming: Expert-level Python . LLM Eco-system: Expertise with Hugging Face , LangChain , AWS/Azure ML and/or LlamaIndex, . Data Engineering: Proficiency with Apache Spark , Airflow , or similar large-scale data processing tools. Vector Databases: Hands-on experience with at least one vector database (e.g., Databases: Strong knowledge of SQL (PostgreSQL) and NoSQL stores. Cloud: High familiarity with either AWS (S3, SageMaker, Glue, EMR) or Azure (Blob, Azure ML, Data Factory). DevOps: Solid understanding of Docker and Git. D. in Computer Science, AI, Data Science, or a related field. Experience building and deploying agentic systems. Direct experience in or a strong understanding of the Venture Capital, Private Equity, or finance domains. Experience working in a startup environment, with a high degree of autonomy and ownership.