Staff Software Engineer, AI

Juniper Square is a company focused on unlocking the potential of private markets through technology, data, and fund administration services. As a Staff Software Engineer, you will lead an engineering team in building an AI-powered document intelligence platform, focusing on transforming unstructured financial documents into structured data and ensuring the reliability of document extraction and retrieval systems.

Responsibilities:

• Champion and embed AI-native development practices and tools (e.g., Cursor, Augment) to achieve significant productivity gains, fostering a "startup-mode" culture of rapid iteration, high velocity, and quality, including guiding the effective use of AI code generation
• Take ownership over the team's architecture, actively participating in design reviews and driving the long-term technical vision
• Take ownership of team code, and actively participate in coding, testing and delivering roadmap projects. Write high-quality code that is well-tested, secure, and maintainable
• Lead the team in designing and implementing document processing pipelines — including extraction, classification, chunking, embedding, and retrieval — and identify opportunities to leverage LLMs and modern AI tooling to improve accuracy, coverage, and scalability
• Own the end-to-end design and reliability of document extraction and RAG pipelines — from ingestion and preprocessing through model inference, post-processing, and structured output delivery. Define quality benchmarks, evaluation frameworks, and feedback loops to continuously improve extraction accuracy
• Effectively manage the team's short-term roadmap (spanning the next two quarters), actively identifying risks and creating clear mitigation strategies to ensure successful project delivery
• Provide backend and AI/ML-focused technical leadership, leveling up existing team members and helping build high-performance teams. Mentor engineers, fostering their technical and professional growth
• Collaborate with cross-functional partners (Product, UX, QA, Customer Support) to ensure the team meets project timelines and solutions align with business strategy. Handle most cross-team conflicts and decisions autonomously
• Own monitoring, diagnosing, and resolving production issues within the team's services
• Contribute directly to efforts through building features and frameworks, conducting code reviews, participating in architecture and system design discussions
• Implement and ensure best practices across the teams to maximize developer productivity
• Actively seek opportunities to improve our platform and developer experience and own those initiatives through execution
• Partner with recruiting to build and grow the team
• Grow into a subject matter expert (SME) in AI document extraction and RAG systems, with a deep understanding of how structured data extraction creates value across the private markets workflow

Requirements:

• Bachelor's degree in Computer Science, Mathematics, AI/ML, or a related technical field
• 7+ years of backend and/or ML engineering experience, with a trajectory of increasing technical leadership, architectural responsibility, and mentorship
• Deep expertise in Python, with strong proficiency in building production-grade backend services and data pipelines; experience with other server-side languages (Node/TS, Java) a plus
• Solid understanding of Python web frameworks (like Django or FastAPI)
• Hands-on experience designing and operating document processing pipelines, including parsing, extraction, classification, and structured output generation from unstructured documents (PDFs, scanned files, financial forms, etc.)
• Production experience building and operating RAG systems, including chunking strategies, embedding models, vector stores (e.g., pgvector, Pinecone, Weaviate), retrieval, and reranking
• Experience evaluating and improving LLM-based extraction quality — including designing eval frameworks, handling edge cases, and building human-in-the-loop feedback mechanisms
• Familiarity with model serving, inference optimization, and managing LLM API costs at scale
• Experience with LLM application patterns beyond basic RAG — including tool-calling agents, planning/execution loops, and multi-step reasoning systems — and how these apply to document intelligence workflows
• Experience with Relational Databases like Postgres or MySQL
• Experience with Cloud technologies (AWS preferred) and Container technologies (Docker and k8s)
• Deep understanding of service-oriented architecture, modern software development practices, and developing scalable, reliable systems
• Ability to identify and evaluate opportunities to integrate AI capabilities into products and workflows
• Demonstrable product focus and a keen understanding of how technology can solve customer problems and drive business outcomes
• Highly self-driven, with a proactive approach to leadership, technical problem-solving, and initiative execution
• Experience working in agile development environments and familiarity with practices that promote rapid iteration and velocity
• Excellent communication and collaboration skills, with the ability to articulate complex technical concepts to both technical and non-technical stakeholders and align them on product goals
• Proven ability to lead projects end-to-end with a player-coach mindset — hands-on in code and architecture while working autonomously with product partners
• Ability to manage multiple priorities and lead teams effectively in a fast-paced environment, with the flexibility to adapt as needs shift
• Demonstrated track record of mentoring engineers and elevating team technical capability
• Hands-on experience with AI-native development tools (e.g., Cursor, Augment, Loveable); demonstrated ability to embed AI-driven practices to accelerate team velocity and code quality
• Ability to critically evaluate AI-generated code and outputs, including identifying failure modes, regressions, and edge cases introduced by AI-assisted development
• Experience building and shipping production-grade software using AI-assisted workflows across the full SDLC
• Experience with OCR technologies and document understanding models (e.g., AWS Textract, Azure Document Intelligence, LayoutLM, Donut)
• Background in financial document processing or fintech data pipelines
• Experience with MLOps tooling (experiment tracking, model registries, deployment pipelines)
• Familiarity with evaluation frameworks for LLM/extraction quality (e.g., RAGAS, custom evals, human review pipelines)
• Experience with multi-modal models or vision-language models for document understanding
• Knowledge of data privacy and compliance considerations in document processing pipelines (PII handling, encryption, access controls)