Job VC
Python Backend Engineer (AI-Heavy SaaS)
Technologies
Description
We’re building AI-driven SaaS platforms that automate and optimize business workflows for SMBs using LLMs, AI agents, and intelligent pipelines.
You’ll work in startup mode, balancing R&D (rapid AI prototyping) and production (scalable, fault-tolerant backend systems).
Your Role
You will own backend services that power AI workflows, agent orchestration, and data pipelines — from experimentation to production scale.
You’ll collaborate closely with AI engineers and product teams to turn models into reliable, monetizable features.
YOUR STACK:
Design and build Python backend services for AI-driven products- Build async, scalable APIs for AI inference and orchestration- Move AI prototypes into production-grade systems.
Ensure observability, performance, and cost efficiency of AI workloads.
Design data pipelines for embeddings, vector search, and context retrieval.
Implement
LLM-powered
features (agents, tools, workflows, RAG).
Backend
Python (FastAPI preferred)
Async IO, background workers (Celery / asyncio / task queues)
REST / event-driven APIs
AI & Data
LLM APIs (OpenAI, Anthropic, open-source models)
Agent frameworks / orchestration (LangChain, custom pipelines)
Vector databases (pgvector, Pinecone, Weaviate, Qdrant)
RAG pipelines, embeddings, prompt engineering
Model inference & evaluation basics
Storage
PostgreSQL
Redis
Object storage (S3-compatible)
Architecture
Service-oriented / microservice design
Event-driven systems (queues, pub/sub)
Cost-aware AI architecture (batching, caching, retries)
NICE-TO-HAVE:
AWS (ECS / EKS / Lambda / RDS / S3)- Experience with high-load or SaaS production systems- Experience deploying or hosting models (vLLM, HuggingFace, inference APIs)
Knowledge of AI observability (tracing, token usage, latency, failure modes)
Docker, CI/CD, infrastructure-as-code
RESPONSIBILITIES:
Design and build Python backend services for AI-driven products- Build async, scalable APIs for AI inference and orchestration- Move AI prototypes into production-grade systems.
Ensure observability, performance, and cost efficiency of AI workloads.
Design data pipelines for embeddings, vector search, and context retrieval.
Implement
LLM-powered
features (agents, tools, workflows, RAG).
OFFER:
Remote work;
Open management without bureaucracy;
Salary reviews according to the results of performance appraisal;
10 days paid sick leave and 18 working days’ vacation;
Days off on National/Bank Holidays according to the legislation of Ukraine;
Real AI in production — not just demos;
Strong R&D culture with ownership;
Opportunity to shape AI backend architecture from day one;
Flexible remote work;
Direct impact & fast feedback loops.
You’ll work in startup mode, balancing R&D (rapid AI prototyping) and production (scalable, fault-tolerant backend systems).
Your Role
You will own backend services that power AI workflows, agent orchestration, and data pipelines — from experimentation to production scale.
You’ll collaborate closely with AI engineers and product teams to turn models into reliable, monetizable features.
YOUR STACK:
Design and build Python backend services for AI-driven products- Build async, scalable APIs for AI inference and orchestration- Move AI prototypes into production-grade systems.
Ensure observability, performance, and cost efficiency of AI workloads.
Design data pipelines for embeddings, vector search, and context retrieval.
Implement
LLM-powered
features (agents, tools, workflows, RAG).
Backend
Python (FastAPI preferred)
Async IO, background workers (Celery / asyncio / task queues)
REST / event-driven APIs
AI & Data
LLM APIs (OpenAI, Anthropic, open-source models)
Agent frameworks / orchestration (LangChain, custom pipelines)
Vector databases (pgvector, Pinecone, Weaviate, Qdrant)
RAG pipelines, embeddings, prompt engineering
Model inference & evaluation basics
Storage
PostgreSQL
Redis
Object storage (S3-compatible)
Architecture
Service-oriented / microservice design
Event-driven systems (queues, pub/sub)
Cost-aware AI architecture (batching, caching, retries)
NICE-TO-HAVE:
AWS (ECS / EKS / Lambda / RDS / S3)- Experience with high-load or SaaS production systems- Experience deploying or hosting models (vLLM, HuggingFace, inference APIs)
Knowledge of AI observability (tracing, token usage, latency, failure modes)
Docker, CI/CD, infrastructure-as-code
RESPONSIBILITIES:
Design and build Python backend services for AI-driven products- Build async, scalable APIs for AI inference and orchestration- Move AI prototypes into production-grade systems.
Ensure observability, performance, and cost efficiency of AI workloads.
Design data pipelines for embeddings, vector search, and context retrieval.
Implement
LLM-powered
features (agents, tools, workflows, RAG).
OFFER:
Remote work;
Open management without bureaucracy;
Salary reviews according to the results of performance appraisal;
10 days paid sick leave and 18 working days’ vacation;
Days off on National/Bank Holidays according to the legislation of Ukraine;
Real AI in production — not just demos;
Strong R&D culture with ownership;
Opportunity to shape AI backend architecture from day one;
Flexible remote work;
Direct impact & fast feedback loops.