AI Engineer (Go & Python) for LLM-Powered Systems

Location: Remote

Company: ShelterZoom

Department: Product & Engineering

About ShelterZoom

ShelterZoom is a deep tech company and multi-category innovator in cybersecurity, digital content ownership, and business continuity. With nearly 80 patents and trademarks in its portfolio, ShelterZoom is redefining data ownership, tracking, and protection.

Recognized by Gartner® as a market leader for three consecutive years (2022–2024), ShelterZoom continues to shape the future of digital resilience. From inventing document tokens and Single Source of Truth® (SSOT) technology to building Spare Tire—a life-saving solution for healthcare EHR downtime—and developing safeguards against AI manipulation, ShelterZoom is creating platforms that define the next generation of data ownership, cybersecurity, and artificial intelligence.

Learn more: www.shelterzoom.com

About the Role

We are looking for an AI Engineer with strong Go and Python expertise to build, deploy, and operate production-grade LLM-powered systems.

At ShelterZoom, we bring together high-performing individuals who think creatively, move fast, and push boundaries. If you are ready to unlock new opportunities, create real-world impact, and help shape the future of digital resilience, we’d love to hear from you.

Key Responsibilities:

• Design, build, and maintain production-ready LLM and RAG-based systems with a strong focus on reliability, scalability, and performance
• Architect and implement Retrieval-Augmented Generation (RAG) pipelines, including hybrid search (dense + sparse), re-ranking, and retrieval validation
• Design and manage LLM memory layers, including:
• Short-term memory (context windows)
• Long-term memory (vector databases)
• Structured memory (key-value stores and relational databases)
• Integrate and optimize LLMs and embeddings for real-world use cases, including chunking, indexing, and retrieval strategies
• Develop and orchestrate AI agents, including multi-agent systems (planner–executor, supervisor–worker patterns)
• Implement robust guardrails and hallucination mitigation techniques using validation, constraints, and monitoring
• Build observability and evaluation pipelines for LLM systems (quality, latency, cost, and safety)
• Optimize systems for latency, cost efficiency, and throughput, including async inference, streaming, caching, and prompt compression
• Collaborate closely with product, security, and infrastructure teams to deliver secure, enterprise-grade AI solutions
• Contribute to system design, code reviews, testing, and continuous improvement of engineering best practices

Requirements:

• Design and implementation of RAG systems, including:
• Hybrid search (dense + sparse)
• Re-ranking strategies
• LLM memory design:
• Short-term (context window)
• Long-term (vector databases)
• Structured memory (KV / relational databases)
• Embeddings & chunking strategies:
• text-embedding-3, BGE, E5
• Sliding window and semantic chunking
• Evaluation frameworks:
• RAGAS, TruLens, DeepEval

• Experience with agent frameworks:
• LangChain, LangGraph
• MCP (Model Context Protocol)
• Multi-agent orchestration patterns
• Prompt engineering:
• System prompts
• Few-shot learning
• Structured outputs

• Guardrails & hallucination mitigation:
• Retrieval validation
• Output constraints
• Observability & monitoring:
• Langfuse, OpenTelemetry, custom tracing
• Cost optimization:
• Token budgeting
• Caching strategies
• Prompt compression
• Latency optimization:
• Async inference
• Parallel calls
• Streaming responses

• Strong proficiency in Python and Go
• Go runtime internals:
• Concurrency model (goroutines, channels)
• Garbage collection
• Core data structures
• Profiling and performance tuning
• Testing practices

Nice to Have:

• Strong understanding of software design principles:
• SOLID, Clean Architecture, GoF patterns
• Idiomatic Go project layout
• Networking fundamentals:
• TCP/IP and core Internet concepts
• Knowledge Graphs / GraphRAG
• Data systems fundamentals:
• OLTP vs OLAP
• ACID, CAP theorem
• Index structures
• Databases:
• PostgreSQL, MySQL, MongoDB (internal principles and components)
• Infrastructure & DevOps:
• Kubernetes, Prometheus, CI/CD, AWS
• Distributed systems:
• Redis, Elasticsearch, Kafka
• Microservices and high-availability systems
• Exposure to additional technologies beyond the core stack
• Russian language proficiency (nice to have, not required)

Job Details

• Remote Policy: Full Remote
• Contract Type: Full-Time, Contract

Benefits

• Competitive salary
• Learning and development opportunities to support individual growth
• Flexible, remote work with a strong focus on work–life balance and wellbeing
• Open-minded, collaborative, and diverse culture
• Clear career growth path in a company with a one-of-a-kind, category-defining product
• Equal opportunities in a respectful, fair, and socially conscious environment
• A people-first culture built on respect and open communication
• Dynamic and flexible international team
• Opportunity to work on cutting-edge AI-driven cybersecurity and digital resilience platforms with proven market recognition