Archer Data Scientist

Archer•Livermore, CA

About The Position

We are seeking an experienced Data Scientist with a strong background in AI model integration, data pipeline development, and knowledge base (KB) engineering to support our next-generation LegalTech / RegTech AI platform. This role blends applied machine learning, data engineering, and software development, focusing on building scalable pipelines that connect large language models (LLMs) to structured and unstructured data through retrieval-augmented generation (RAG) and vector database architectures. The ideal candidate is passionate about operationalizing AI — from training and fine-tuning models to deploying intelligent retrieval systems in AWS cloud environments.

Requirements

5+ years of experience in data science, ML engineering, or AI-driven software development.
Strong programming skills in Python (NumPy, Pandas, PyTorch/TensorFlow, LangChain, or equivalent).
Experience with vector databases and retrieval systems (Pinecone, FAISS, Weaviate, Qdrant, or Elastic Vector Search).
Hands-on experience with RAG pipelines, embedding models, and LLM orchestration (OpenAI, Bedrock, Hugging Face, etc.).
Solid understanding of data pipelines, ETL frameworks, and cloud-native deployment on AWS.
Familiarity with Elasticsearch, PostgreSQL, and API integration patterns.
Knowledge of ML lifecycle management, including model training, evaluation, and monitoring.
Strong problem-solving and system design capabilities.
Excellent communication skills for cross-disciplinary collaboration.
Passion for structured documentation, reproducibility, and experimentation.
Adaptable mindset with focus on performance, scalability, and reliability.
All employees must be legally authorized to work in Country they are applying for.

Nice To Haves

Experience building AI products for LegalTech, RegTech, or compliance automation.
Familiarity with agentic AI frameworks (e.g., OpenAI MCP, CrewAI, LangGraph, or AutoGen).
Background in document intelligence systems, multi-agent orchestration, or knowledge graph integration.
Experience with LangChain, LlamaIndex, or similar frameworks for RAG orchestration.
Hands-on knowledge of MLOps tools and data versioning (DVC, MLflow, Weights & Biases).
Understanding of governance, interpretability, and ethical AI.

Responsibilities

Design, train, and evaluate LLM-based pipelines for document understanding, obligation extraction, and regulatory reasoning.
Implement and optimize RAG architectures, combining LLMs with vector databases for semantic retrieval.
Develop and maintain model fine-tuning workflows, embedding generation, and knowledge distillation.
Collaborate with ML Ops teams to integrate AI models into production-ready APIs and services on AWS.
Measure and improve model precision, recall, latency, and interpretability.
Design and maintain agentic multi-component processes (MCPs) that enable context-aware reasoning across multiple data sources and agents.
Implement AI agents capable of dynamic tool use, autonomous task decomposition, and multi-context knowledge retrieval.
Develop pipelines that support agent memory, self-reflection, and knowledge synthesis across distributed systems and knowledge bases.
Collaborate with engineering teams to integrate MCP-driven agents with retrieval, analytics, and workflow orchestration layers, ensuring compliance with regulatory reasoning frameworks.
Build and manage end-to-end data pipelines for ingestion, transformation, embedding, and indexing of legal and compliance data.
Orchestrate data workflows leveraging AWS services (e.g., S3, Lambda, Glue, SageMaker, Step Functions, RDS).
Develop scalable ETL/ELT processes to feed both relational (PostgreSQL) and vector databases (e.g., Pinecone, FAISS, Weaviate, Elastic Vector Search).
Ensure data lineage, reproducibility, and version control across AI and analytics pipelines.
Automate retraining and evaluation pipelines for continuous learning from user feedback.
Architect and maintain intelligent Knowledge Bases (KBs) to support AI-driven search, summarization, and compliance reasoning.
Implement advanced retrieval techniques using ElasticSearch / Elastic Vector Search and embedding-based retrieval.
Align KB structures with business ontologies and regulatory taxonomies to support explainable AI outputs.
Collaborate with domain experts and PMs to enrich KB metadata and enhance model context relevance.
Deploy and scale AI pipelines using AWS services such as SageMaker, Lambda, ECS/EKS, API Gateway, and CloudFormation/Terraform.
Implement model and data monitoring solutions for drift detection, latency management, and cost optimization.
Collaborate with DevOps to maintain secure, reliable, and compliant cloud environments.
Partner with engineering, product, and compliance teams to align AI models with regulatory and data governance requirements.
Work closely with QA and Professional Services teams to validate AI outputs and improve client-facing performance.
Document architectures, experiment results, and data flows to ensure transparency and reproducibility.