Are you interested in building production-grade AI systems that go beyond demos and simple LLM wrappers?
About the product
At Forma Pro, we work on AI-powered products and operational systems for companies across marketplaces, MarTech, PropTech, logistics and other data-intensive industries.
Our projects combine LLMs, agentic workflows, structured and unstructured data, external APIs and existing business systems. We focus on building reliable AI solutions that can operate in real production environments, where evaluation, observability, security and predictable behaviour matter as much as model capabilities.
Technologies that we use
- Python
- FastAPI
- OpenAI, Anthropic and open-source LLMs
- LangGraph and agent orchestration frameworks
- RAG and vector search
- PostgreSQL and pgvector
- Redis
- AWS
- Docker
- Modern observability and evaluation tooling
We do not expect you to have worked with every tool on this list. We care more about strong engineering fundamentals and your ability to select the right technology for a problem.
About you
- You are interested in building AI systems that people actually use in production
- You think beyond prompts and consider the whole system: data, tools, state, evaluation, reliability and user experience
- You are comfortable working independently and making technical decisions
- You are willing to research and test new approaches instead of blindly following framework conventions
- You care about measurable quality rather than impressive-looking demos
- You have a proactive approach and are comfortable taking ownership of a problem from idea to production
Required skills and competencies
- Spoken English — upper-intermediate or above
- 4+ years of software engineering experience
- Strong Python development skills
- Hands-on experience building applications with LLM APIs
- Experience with RAG, embeddings, semantic search or similar retrieval systems
- Understanding of agentic architectures, tool/function calling and multi-step workflows
- Experience integrating external APIs and business systems
- Solid understanding of backend architecture, databases and asynchronous processing
- Experience writing automated tests and working with production systems
Would be a plus
- Experience designing LLM evaluation and regression testing pipelines
- Experience with LangGraph, PydanticAI, Semantic Kernel or similar frameworks
- Experience with multi-agent systems
- Experience with model routing, structured outputs and guardrails
- Experience optimizing AI systems for latency and inference cost
- Experience with self-hosted or open-source models
- Familiarity with MCP and emerging agent interoperability standards
Responsibilities
- Design and build production-grade AI applications and agentic workflows
- Develop systems that combine LLMs with APIs, databases, internal tools and business processes
- Design RAG and knowledge retrieval architectures for structured and unstructured data
- Build evaluation pipelines and regression tests for prompts, retrieval and agent behaviour
- Design error handling, fallbacks and human-in-the-loop workflows for situations where AI should not make an autonomous decision
- Improve latency, reliability and cost efficiency of AI systems
- Research new models, frameworks and techniques and determine whether they are useful in real products
- Participate in architecture decisions together with backend, data and product engineers
- Take ownership of features from initial technical design through deployment and iteration
What we offer
- No micromanagement — we do not track your working hours, breaks or keyboard activity
- Freedom to choose how you solve a problem — we care about the result and engineering quality
- Flexible working hours
- Fully remote work — you can work from almost anywhere
- Direct influence on architecture and technical decisions
- An opportunity to work with modern AI technologies on real production systems rather than endless proof-of-concepts
- Company support when you need it
- Direct communication with Forma Pro's CEO and CTO