ABOUT CORPUS

Corpus is an Algerian LegalTech startup building AI-powered products that help legal professionals work with complex legal information more efficiently.

Our flagship product, ScribeAI, is live in the market and already trusted by leading organizations. It's an advanced legal AI assistant that transforms how legal professionals search, analyze, and work with legal information.

We're looking for a Backend Engineer to join our engineering team, help maintain, improve, and extend the AI workflows behind ScribeAI, and contribute to future Corpus products. Around 2 years of experience is preferred, but strong junior candidates are welcome.

ABOUT THE ROLE

You'll work on the backend behind ScribeAI, a live production application built around LangChain, LangGraph, multiple LLM providers, and a combination of databases and retrieval systems. This isn't a demo or a prototype. It's a real product with real users, and your work directly affects the accuracy and reliability of the answers it gives. You'll improve existing capabilities, investigate production issues, deliver new features, and make our AI systems more accurate, efficient, and maintainable.

You won't be doing it alone. You'll join a team that already knows the system, shares the workload, and will help you get up to speed.

We're looking for someone who does more than connect an API to a chatbot, and who is curious about what happens behind the scenes. How do we retrieve the right information? How do we prevent unsupported answers? How do we make AI workflows reliable enough for professional legal use? You'll work on these questions directly, alongside frontend engineers and the rest of the engineering team.

WHY JOIN CORPUS

Legal technology is a new and largely untapped field in Algeria, and ScribeAI is already proving its value across the public and private sectors.

A more advanced product is in the works, and the team is growing behind a roadmap with a clear business strategy. Joining Corpus means helping define what LegalTech looks like here.

WHAT YOU'LL DO

• Maintain and improve the LangChain and LangGraph workflows that route legal questions, coordinate agents, and synthesize answers
• Improve query routing, agent coordination, and multi-step research workflows such as Deep Search, and investigate incorrect or inconsistent AI responses
• Maintain the web search and crawling workflows used for research
• Connect our workflows to the retrieval layer (vector search and knowledge graph): query it, inspect results, and improve relevance, contextual grounding, and citation traceability
• Integrate and maintain LLM providers, with structured outputs, validation, retries, and fallbacks
• Integrate third-party document extraction and analysis services (LLM-based or other) through their APIs, and feed their output into our document analysis workflows
• Maintain document upload and export APIs
• Develop backend services and APIs, including authentication and reliable asynchronous and streaming workflows
• Work with MongoDB for conversations, sessions, agent state, and application data
• Work with frontend engineers on API contracts and streaming event schemas
• Diagnose and resolve production issues, and improve performance, reliability, and observability
• Write tests for backend services and AI workflows, and help evaluate retrieval quality and generated responses
• Contribute to reusable backend and AI components for future Corpus products

TECH STACK

• Languages & backend: TypeScript, Node.js, Next.js API routes, streaming APIs
• AI orchestration: LangChain, LangGraph, Zod (structured outputs), multiple LLM providers
• Databases & retrieval: MongoDB (Mongoose), Neo4j, ChromaDB, vector embeddings
• Web search & crawling: Tavily
• Authentication: Clerk
• Document analysis: third-party extraction and analysis APIs
• Testing & deployment: Vitest, Docker

WHAT WE'RE LOOKING FOR

• Around 2 years of relevant professional experience is preferred. Candidates who show equivalent ability through substantial projects, internships, open-source contributions, or demonstrated experience building AI systems are equally welcome
• Good programming skills in TypeScript, or in another language with the ability to work in a TypeScript/Node.js backend
• Solid understanding of backend development, APIs, and asynchronous programming
• Experience integrating third-party APIs and services
• Familiarity with LLM APIs and the fundamentals of building AI-powered applications
• Understanding of RAG, embeddings, semantic search, or information retrieval
• Familiarity with databases and data modeling
• Ability to debug existing code and improve systems without introducing regressions
• Clear communication and the ability to work closely with frontend and product teams
• A proactive mindset, willingness to learn, and a sense of ownership over your work

NICE TO HAVE

• Hands-on experience with LangChain or LangGraph
• Experience building RAG pipelines or LLM-powered applications
• Familiarity with Neo4j, Cypher, or knowledge graphs
• Experience with MongoDB, ChromaDB, or other vector databases
• Familiarity with structured LLM outputs, tool calling, retries, and model fallbacks
• Experience with document extraction or OCR services, or with unstructured data
• Experience evaluating AI systems: retrieval testing, citation verification, or answer quality
• Familiarity with Docker, cloud deployment, or observability tools
• Experience with Python, multilingual applications, or legal information systems

WHAT MATTERS MOST

You don't need to be an expert in every part of our stack. We value strong fundamentals, practical projects, systematic debugging, sound engineering judgment, and the ability to learn quickly. If you understand how information is retrieved, how failures are handled, and how the quality of AI-generated answers can be improved, we want to hear from you.

WHAT WE OFFER

• Competitive compensation that grows with your tenure and your impact
• A clear growth path: engineers who prove themselves can move into a team lead role as the team expands
• Responsibility that grows with your performance, from improving existing systems to shaping how new ones are built
• Work directly with the CTO and a multidisciplinary engineering team
• Hands-on experience with production AI systems: multi-agent workflows, RAG, and document analysis

HOW TO APPLY

Apply directly through this LinkedIn posting. For the fastest review, include links to your GitHub or a project you've built, ideally involving AI, retrieval, or backend systems, with a short description of what it does.