Senior AI/LLM Backend Engineer
Job Details
About the Company
With operational hubs scattered across Europe, Asia, and LATAM, and its headquarters situated in San Francisco, US, the company boasts a workforce of over 1,000 adept professionals. Spanning across more than 20 countries, ALLSTARSIT offers a diverse range of skilled employees across various verticals, including AI, cybersecurity, healthcare, fintech, telecom, media, and so on.
About the Project
The platform is a unified advertising technology stack that connects advertisers, publishers and data partners across digital and connected TV environments. It enables buyers to plan, activate and measure full-funnel campaigns, while helping publishers maximize inventory and data monetization.
The platform combines a demand-side platform, supply-side platform and proprietary data capabilities, including ACR data and AI-driven optimization, to deliver measurable outcomes across screens for both brand and performance advertisers.
Specialization
Headquarters
Years on the market
Team size and structure
Current technology stack
Required skills:
Must-have skills:
- Strong Python backend engineering experience (FastAPI as primary framework)
- Hands-on experience building LLM-powered workflows and agentic systems
- Experience with LLM orchestration frameworks: LangChain and/or LangGraph
- RAG architectures knowledge
- Prompt engineering experience
- Experience building and evaluating AI pipelines in production
- Relational DB experience: PostgreSQL
Nice-to-have skills:
- Node.js / NestJS (secondary backend - advantage, not required)
- Non-relational DB experience: vector DB or graph DB (meaningful advantage)
- Message queue experience: Kafka or Celery
- Redis (caching, LLM response handling)
- AWS cloud environment
- Docker + Kubernetes (containerization)
- LLM pipeline evaluation experience
- AdTech / DSP / SSP domain background
- Mentoring experience (junior/mid-level developers)
Scope of work:
- Building and maintaining AI workflows across three product domains
- Designing and implementing agentic pipelines and RAG architectures
- Prompt engineering and LLM integration into production Python services
- Evaluation of LLM pipelines as part of day-to-day work
- Context-switching between different product areas and user types
- Engaging stakeholders with varying levels of AI maturity across the organization
- Writing and maintaining automated tests
- Collaboration with cross-functional, internationally distributed teams (US + Israel)
- Active use of AI tools as part of daily workflow