Software Engineer – Backend & AI Agents

Filled
February 28, 2026
Apply Now

Job Description

Company: Recurrent Software
Location: Remote (100%)
Job Type: Full-Time
Experience: 3+ years

🔹 Company Overview

Recurrent Software is a fast-moving startup leveraging Generative AI and Machine Learning to build intelligent products for real-world problems. The company focuses on creating AI-powered autonomous agents, scalable backend systems, and innovative workflows that deliver measurable impact across industries.

🔹 Role Overview

This is a 50/50 split role between:

  1. Backend Engineering – designing robust, scalable backend services.
  2. AI & LLM Agent Engineering – building autonomous AI agents using LLMs and vector databases.

You will work closely with the founding team, shaping product architecture and engineering culture from day one.

🔹 Backend Engineering Responsibilities (50%)

  • Develop and maintain RESTful APIs using Python and FastAPI with high performance and reliability.
  • Architect relational and NoSQL database schemas (PostgreSQL, MongoDB, etc.) including indexing, migrations, and query optimization.
  • Implement authentication, authorization, rate-limiting, and other API security best practices.
  • Write tests (unit, integration, end-to-end) and maintain CI/CD pipelines.
  • Build scalable microservices and background tasks using Celery, Redis, or message queues.
  • Collaborate on infrastructure: containerization (Docker), orchestration (Kubernetes), and cloud services (AWS/GCP/Azure).
  • Maintain well-structured, reviewable code and clear API documentation (OpenAPI/Swagger).

🔹 AI & LLM Agent Engineering Responsibilities (50%)

  • Design and develop autonomous AI agents using frameworks like LangGraph, LangChain, CrewAI, AutoGen.
  • Integrate LLMs (OpenAI, Anthropic, open-source models) with production workflows, including prompt engineering, function/tool calling, and output parsing.
  • Build RAG pipelines: document ingestion, chunking, embedding generation, and semantic search.
  • Work with vector databases (Pinecone, Weaviate, Qdrant, ChromaDB, Milvus).
  • Implement agent memory, multi-step reasoning, planning loops, and tool-use patterns.
  • Evaluate, benchmark, and iterate on LLM outputs and monitor performance/cost.
  • Stay up-to-date with AI/LLM technologies and proactively suggest improvements.

🔹 Required Qualifications

  • 3+ years of professional backend/software engineering experience.
  • Strong proficiency in Python and FastAPI (or Flask/Django).
  • Solid knowledge of relational databases (PostgreSQL/MySQL) and NoSQL databases (MongoDB/DynamoDB).
  • Hands-on experience with LLM-based applications or AI agents.
  • Working knowledge of vector databases (Pinecone, Weaviate, Qdrant, ChromaDB, or pgvector).
  • Understanding of embeddings, semantic search, and RAG architecture.
  • Comfortable in a fast-paced, fully remote startup environment.
  • Strong communication skills and ability to work asynchronously.

🔹 Nice-to-Have

  • Multi-agent orchestration and complex agentic workflows.
  • Fine-tuning/training LLMs and working with open-source models (LLaMA, Mistral).
  • Experience with graph databases (Neo4j) or knowledge graph construction.
  • Event-driven architectures, streaming data, or real-time systems.
  • MLOps experience: model deployment or inference optimization.
  • Contributions to open-source AI/ML projects.

🔹 What We Offer

  • Competitive salary + equity in an early-stage startup.
  • Fully remote with flexible hours and async-first culture.
  • Significant influence on architecture, product direction, and engineering practices.
  • Small, high-impact team with meaningful ownership.
  • Learning budget and conference allowance.
  • Generous PTO and culture valuing sustainable pace.