We are looking for a senior ai/full-stack engineer or senior team to develop an ai agent platform built with deep agents, langchain and langgraph.
Our goal is to create a scalable B2B SaaS with Super Agents capable of completing complex, long-running missions, coordinating subagents, maintaining persistent memory and context, using APIs, MCP and browser automation, recovering from failures, and operating under configurable autonomy and human approval.
The system must be advanced but simple for end customers. It will include a dashboard for managing agents, skills, models, knowledge, memory, integrations, tasks, approvals, costs, permissions, audit logs and automations without changing code during normal operations.
Stack: Python, Deep Agents, LangChain, LangGraph, FastAPI, PostgreSQL/pgvector, MCP, Playwright/Browser Use, Docker,
Next.js and TypeScript.
We already have a technical specification. Development will be contracted in phases:
1. Technical audit and architecture validation.
2. Super Agent proof of concept.
3. Private MVP with administration dashboard.
4. Reproducible deployment on vps/aws.
5. Internal validation, security and recovery.
6. Deployment for 2 to 5 pilot customers.
7. Evolution into a multi-tenant SaaS.
8. White-label and marketplace.
We require experience with Python, LangChain/Langgraph, ai agents, durable workflows, mcp, browser automation, postgresql, docker, security and saas architecture.
All code must be stored in our company GitHub, with tests and documentation. Original code, dashboards, workflows, skills and integrations developed for the project must be assigned to our company, while respecting open-source licences.
We request a fixed quotation for phases 1 and 2 and an estimate for the remaining phases. Please send relevant experience, portfolio/GitHub, availability, timeline and price per phase. We are not looking for a chatbot, but for an autonomous, durable and auditable enterprise agent platform.
Prazo de Entrega: Não estabelecido