Backend and integration
Modern Python, asynchronous programming, APIs and web services, data modeling and systems integration. We mainly use FastAPI and Pydantic.
Professional collaborations in Python software development, AI and data.
Fully remote
At Isagog Srl we make organizations' knowledge accessible to people and applications. We combine language models, knowledge graphs and reasoning to build AI systems whose answers are grounded in verifiable information and whose limits are recognizable.
Our work starts from users' problems and ends in software that helps them tackle those problems. We measure a solution's quality by its usefulness, by how well its behavior can be understood, and by the care with which it is maintained.
We are a small company where research and development work side by side. We are looking for people who want to contribute to advanced AI projects, take ownership of concrete assignments and work alongside professionals with proven academic and research experience.
Backend and the components that connect data, knowledge and AI agents.
You will work mainly on the backend and on the components that connect data, knowledge and AI agents. Depending on your skills and the project, you may build services and APIs, integrate information sources, build processing pipelines or improve the reliability and evaluation of our systems.
We expect you to follow a task all the way through: understand the need, clarify the constraints, propose a solution, build it, verify it and document its delivery. Architectural choices must have an understandable rationale and keep the software tidy, easy to change and proportionate to the problem.
We are looking for a solid foundation in Python software engineering, together with hands-on experience in one or more of the AI and data areas we work in:
Modern Python, asynchronous programming, APIs and web services, data modeling and systems integration. We mainly use FastAPI and Pydantic.
Language models and agents, semantic search and RAG, evaluating results. Knowledge graphs, ontologies and RDF/SPARQL are a specialization that matters especially to us.
Relational and vector databases, pipelines and workflows, error handling and restarts. Our stack includes PostgreSQL/pgvector, Redis and Temporal.
Git, automated tests, technical documentation, Docker and familiarity with Linux environments and continuous integration.
Being able to find your way around a TypeScript front end is useful. Optional skills we value include experience with React, English, and the ability to lead presentations and demos, explaining solutions and results clearly to non-technical audiences too.
You don't need to know every tool already: in your application, tell us where you work independently and what you'd like to learn more about. We assess the quality of your work and your ability to learn through concrete examples.
The work is fully remote and calls for autonomy, responsibility and collaboration. For us, that means:
Understanding the user's point of view, asking the right questions and turning needs and constraints into shared requirements and acceptance criteria.
Clarifying the approach before building, discussing alternatives and trade-offs, and documenting the reasons behind decisions.
Agreeing on goals and timelines, organizing the work, verifying the result and taking care of the delivery.
Communicating clearly, flagging problems and dependencies early, and asking for a discussion when needed.
We are looking for the ability to use coding assistants such as Codex, Claude Code or equivalent tools professionally. For us, that means providing the repository context, clear specifications and design constraints, and steering the work toward targeted changes that fit the architecture, structure and conventions of the existing software.
We expect you to choose which tasks to hand to assistants, split the work into verifiable steps and review proposals critically. Generated code must be understood, reviewed and verified with relevant tests, also checking that it preserves each component's responsibilities and adds no duplication or unnecessary complexity.
Technical responsibility for the choices and the delivery stays with the developer.
We offer assignments on real advanced-AI projects, with a recognizable contribution to the product and direct exchange with experienced people from research, academia and complex-systems development.
The scope of the work, the expected results and how we coordinate will be agreed before you start.
Subject line: «Python / AI collaboration – First name Last name»
Include:
If you have code, a portfolio or documentation you can share, add the links. You can also describe a non-public project, respecting its confidentiality.
Send your application ↗In the interview we'll also dig into how you use coding assistants: which tools you choose, how you steer them and at which stages you use them, from analysis to design, development, testing and review. We'll ask for concrete examples, including times you corrected or discarded a proposal because it didn't respect the requirements or the architecture.
We want to understand how you approach problems, how you collaborate and what you can bring to Isagog.