We are building a data-driven B2B SaaS dashboard and need a Senior Full-Stack Engineer to lead a focused 4-week sprint. The core goals for this sprint are: Implementing a modular scoring engine for companies/users based on multiple business rules and signals. Integrating Large Language Models (LLMs) to enrich data, summarize insights, and assist end users inside the dashboard. Connecting to our existing data sources and exposing the scores and insights in a clean, performant frontend dashboard. You will work closely with the founder/product lead and another engineer, with clear weekly milestones and deliverables. Tech stack (must-have experience) Backend: Python or Node.js (TypeScript preferred), working with REST/GraphQL APIs. Frontend: React and/or Next.js for a modern SPA dashboard. Database: PostgreSQL (designing schemas, writing efficient queries, basic optimization). LLMs: Experience integrating APIs such as OpenAI, Anthropic, etc. (prompt design, handling latency/cost, basic evaluation). SaaS + architecture: Comfortable designing modular, evolvable services (e.g., scoring engine that changes over time). Nice to have Experience with data-driven SaaS dashboards for B2B. Background in analytics, scoring, or rule-based decision systems. Experience with background jobs/queues and basic observability (logging, monitoring). What you’ll do (4-week sprint) Design and implement a scoring engine that can be extended with new rules and weights over time. Integrate LLM APIs into our backend and surface results into the dashboard. Implement or refine the dashboard UI for scores, filters, and detail views. Collaborate on architecture decisions, document your work, and hand off clean, maintainable code. Requirements Strong full-stack background (you’ve shipped production SaaS before). Hands-on experience integrating LLM APIs in production systems. Solid understanding of PostgreSQL and data modeling. Good communication, availability for async updates and occasional calls. To apply, please answer briefly: Describe a SaaS platform you built that involved scoring, evaluation, or rule-based logic. Share 1–2 examples where you integrated LLM APIs (OpenAI, Anthropic, etc.) into production. How would you design a modular scoring engine that must evolve over time (high-level approach)? What is your weekly availability over the next 4 weeks (approx. hours/week)?