Own the architecture for our hybrid search engine, blending keyword-based retrieval with dense vector embeddings to improve precision and recall.
Design and scale personalization algorithms that suggest products based on project specs, historical data, and cross-catalog compatibility.
Lead the fine-tuning of open-source and proprietary LLMs/encoders for specialized construction domain tasks, including NER and relationship extraction from complex documents.
Architect and optimize our vector database strategy for high-concurrency retrieval and low-latency ranking.
Work closely with product managers, UX designers, and business leadership to integrate AI components into fully functional systems.
Participate in the complete product lifecycle from concept design to development, testing, and deployment.
Build products that handle large data volumes efficiently while remaining highly scalable for new clients.
Design end-to-end data and ML pipelines for seamless production integration and monitoring.
Work with the leadership team on research efforts to explore cutting-edge technologies.
Uphold a culture of excellence by maintaining high standards in code quality, innovation, and rigorous experimentation.

Requirements

Bachelor’s or Master’s degree (PhD preferred) in Science or Engineering with strong programming and analytical skills.
Deep conceptual understanding and hands-on experience in Search, Ranking, Recommendation systems, or NLP/Document Extraction.
Expertise in Python (NumPy, scikit-learn, pandas) and training deep learning models using PyTorch or TensorFlow.
Ability to drive high standards for clean, efficient, and bug-free code.
Deep experience with Learning to Rank (LTR), BM25, and hybrid retrieval strategies.
Hands-on experience with Vector Databases (Pinecone, Qdrant, Milvus) and optimizing embedding spaces for domain-specific retrieval.
Expertise in fine-tuning Large Language Models (LLMs) and Bi-Encoders/Cross-Encoders for specialized semantic search.
Experience building evaluation frameworks for search (nDCG, MRR) and managing the lifecycle of embedding deployments.
Hands-on experience with agentic frameworks (e.g., LangGraph, AutoGen, or CrewAI) for building complex, multi-step reasoning chains.
A track record of publications in top-tier conferences (e.g., NeurIPS, SIGIR, KDD, ACL) or significant contributions to open-source ML projects.

Benefits

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills & Tools

PythonNumPyscikit-learnpandasPyTorchTensorFlowLearning to RankBM25NLPDocument Extraction

Soft Skills

analytical skillsleadershipcommunicationcollaborationinnovationcode qualityproblem-solvingresearchhigh standardsrigorous experimentation