Design and build core backend services for AI/ML runtimes, implement retrieval and memory systems, and collaborate with ML teams on technology decisions.
REQUIREMENTS:
- Total experience of 6 years+
- Strong expertise in Python and backend engineering with experience building scalable, distributed microservices.
- Hands-on experience designing and delivering end-to-end RAG (Retrieval-Augmented Generation) workflows in production systems.
- Solid understanding of ML solution design, including embeddings, retrieval, ranking, feature engineering, and evaluation strategies.
- Experience with vector databases (FAISS, Pinecone, Milvus, Weaviate) and implementing chunking, indexing, vector search, re-ranking, caching, and memory patterns.
- Knowledge of LLM/NLP engineering, including prompt engineering, model integration, orchestration tools (LangChain/LlamaIndex), and evaluation instrumentation.
- Experience productionizing ML systems with observability, online/offline parity, and performance optimization across latency, throughput, and cost.
- Strong backend integration skills using REST/gRPC APIs, Docker, Kubernetes, CI/CD, and cloud platforms (AWS/GCP/Azure).
- Ability to independently design, ship, and operate reliable, scalable, and cost-efficient ML-backed backend systems with strong ownership mindset.
RESPONSIBILITIES:
- Design and build core backend services powering AI/ML runtime including orchestration, session/state management, and tools/services integration.
- Implement end-to-end retrieval and memory systems covering ingestion, embeddings, indexing, vector search, ranking, caching, and lifecycle management.
- Productionize ML workflows with feature/metadata services, model integration contracts, and evaluation hooks.
- Drive performance, reliability, and cost optimization with strong SLO ownership and observability practices (logs, metrics, tracing, guardrails).
- Collaborate with applied ML teams on model routing, prompts/tools, evaluation datasets, and safe releases.
- Translate business requirements into scalable technical designs, define NFR benchmarks, and review architecture for extensibility and best practices.
- Lead troubleshooting, root-cause analysis, and POCs to validate technology and design decisions.
Bachelor’s or master’s degree in computer science, Information Technology, or a related field.
