SpotDraft provides a privacy-first, on-device AI platform that automates contract review and broader contract lifecycle management for legal and regulated enterprise teams.
ABOUT SPOTDRAFT SpotDraft is on a mission to help legal and business teams move faster, together. Our AI-powered contracting platform is redefining how companies manage contracts, and our story deserves to be told in creative, human, and memorable ways. JOB SUMMARY SpotDraft is revolutionizing legal operations with AI-powered tools that help legal teams work faster and smarter. Our flagship products leverage cutting-edge LLM technology to automate routine legal work and accelerate contract review. Sidebar is an AI-powered team of legal assistants that handles routine legal work—from contract analysis and legal research to compliance tracking and drafting SpotDraft https://www.spotdraft.com/sidebar. It tackles everything beyond contracts, including policy questions, regulatory compliance, and strategic advice SpotDraft https://www.spotdraft.com/blog/sidebar-our-vision-for-legal-work-beyond-contract-management, learning from your organization's knowledge to become a specialized legal co-pilot. VerifAI is an AI contract review tool that works as a Microsoft Word add-in, helping legal teams review contracts up to 70% faster with SpotDraft. It uses generative AI to check contracts against personal or organizational guidelines and answer open-ended questions SpotDraft https://www.spotdraft.com/blog/introducing-verifai-by-spotdraft, automatically flagging deviations and suggesting improvements. THE ROLE As a Junior Applied AI Engineer, you'll help build and maintain the production AI systems powering our legal-tech products. Working alongside senior engineers, you'll contribute to distributed AI services that process large volumes of legal documents, support multi-agent architectures, and help optimize LLM performance for accuracy, speed, and cost in our mission-critical SaaS environment. WHAT YOU'LL DO Build AI Features - Implement and iterate on AI features using transformer-based architectures and prompt engineering for legal document workflows (summarization, clause extraction, document comparison, drafting assistance) - Contribute to agentic systems (ReAct-style flows, tool/function calling, multi-turn reasoning) under the guidance of senior engineers - Write and refine prompts, and help maintain prompt libraries and versioning Support RAG & Context Systems - Help build and maintain retrieval pipelines: chunking strategies, hybrid search (BM25 + dense embeddings), and reranking for legal documents - Assist in building retrieval-grounded generation pipelines and tuning context window usage - Work with vector databases (Pinecone, Weaviate, Qdrant, or pgvector) to support semantic search features Contribute to Infrastructure & Performance - Help build and maintain API services that call LLM providers (OpenAI, Anthropic, Google Gemini), including retry logic, rate limiting, and fallback handling - Support inference optimization efforts: caching, batching, and monitoring latency/cost - Write clean, tested, production-grade Python code within existing service architectures Support Quality & Reliability - Help build evaluation scripts and test datasets to measure model output quality (accuracy, hallucination rate, relevance) - Assist with LLM-as-judge pipelines and human-in-the-loop labeling workflows - Participate in A/B testing of prompts, models, and configurations MLOps & Observability - Support CI/CD pipelines for AI feature deployment (containerization, basic Kubernetes usage) - Help maintain monitoring dashboards for latency, token usage, error rates, and cost per request - Assist in debugging production issues using logs, traces, and dashboards WHAT WE'RE LOOKING FOR Must Have - Upto 3 years of experience in software/ML engineering, with at least some hands-on exposure to LLM-based systems - Working experience integrating LLM APIs (GPT-4, Claude, Gemini, or Llama) — prompt engineering, function calling, or basic fine-tuning - Know-how of context engineering — structuring, trimming, and managing what gets passed into the model's context window across a workflow - Understanding of agent loops and graphs — how agentic control flow works (e.g. ReAct-style loops, state graphs, conditional branching between steps/tools) - Hands-on experience building agentic workflows — chaining tool calls, managing state across turns, and handling multi-step reasoning - Experience building or contributing to eval systems — writing test cases, scoring outputs, and tracking quality/regression over time - Solid Python skills with an understanding of core software engineering practices (testing, version control, code review) - Foundational understanding of transformer architectures, embeddings, and how LLMs work - Some exposure to vector databases and semantic search concepts - Familiarity with async programming and API frameworks (FastAPI or similar) - Basic understanding of data structures, algorithms, and complexity - Comfort working in a fast-paced environment and learning quickly from senior engineers and code reviews Good to Have - Exposure to agentic frameworks (LangChain, LangGraph, LlamaIndex, CrewAI, or similar) - Exposure to text extraction and text wrangling in current/prior work (parsing PDFs, OCR output cleanup, handling messy or unstructured text) - Experience with Docker and basic CI/CD pipelines - Familiarity with cloud platforms (AWS, GCP, or Azure) - Any experience with evaluation frameworks, synthetic test data, or A/B testing for ML systems - Interest in or coursework related to document/multi-modal AI (OCR, layout analysis, vision-language models) - A personal project, hackathon, or open-source contribution involving LLMs or RAG systems GROWTH PATH This role is designed to grow into a mid-level/Applied AI Engineer position, with mentorship from senior team members on distributed systems design, advanced RAG architecture, and large-scale inference optimization. WHY SPOTDRAFT? - Brilliant teamm