Overview

Member of Technical Staff sought by Tungsten Dev, Inc., Mountain View, CA, to perform the following duties: 1. Designing, developing, and implementing backend services and APIs for an enterprise AI agent orchestration platform using Golang, gRPC, and REST APIs to support reliable, high-throughput execution of LLM-driven agentic workflows for Fortune 500 customers. 2. Architecting and validating high-performance storage and retrieval layers for agent memory, context, and knowledge-base grounding using the Milvus vector database and related embedding-based retrieval systems to power retrieval-augmented generation (RAG) within LLM-driven AI agents. 3. Designing and deploying stateful, long-running AI agent workloads on AWS, including configuring Kubernetes (k8s) persistent volumes backed by NFS and S3 protocols as the primary storage layer for durable agent session state, checkpointing, and on-demand replication across availability zones. 4. Implementing secure inter-service communication across the distributed agent platform by integrating the Go-spiffe library with gRPC to enforce mTLS-based identity and authorization between agent runtimes, tool servers, and MCP (Model Context Protocol) proxies. 5. Building and operating containerized AI microservices including model-invocation gateways, MCP proxy layers, and webhook-driven event processors using FastAPI, Docker containers, and GPU-optimized data pipelines deployed on AWS ECS and AWS Bedrock AgentCore. 6. Performing coding, debugging, and testing of software components across the full stack, including agent runtime services, backend APIs, model-routing layers, and deployment infrastructure on AWS (Bedrock AgentCore, ECS, S3, Terraform). 7. Contributing to the design and development of product features as part of a cross-functional engineering team, including OAuth-based third-party integration abstractions, MCP tool proxies, and webhook-driven event triggers for AI agents. 8. Collaborating with enterprise customers and internal teams to gather requirements, translate business needs into technical designs, and implement features for AI-powered workflow automation. 9. Troubleshooting and resolving technical issues related to AI-powered application functionality and enterprise data integrations, including applying Python, PyTorch, and auto-encoder-based anomaly detection pipelines to monitor agent behavior and identify abnormal execution patterns, and documenting code, processes, and technical specifications to support team knowledge sharing and product maintenance. Education and Experience Requirements This position requires a Master's (or foreign educ. equiv.) Degree in Computer Science, Software Engineering, Computer Engineering or a closely related field plus one (1) year of experience in the job offered or a related occupation. Special Skills Requirements Experience must include: a. Designing and implementing replication engine using Golang, gRPC, REST APIs. b. Designing and validating high-performance unified storage architecture using Milvus vector database. c. Adding support for files as the primary storage for k8s persistent volume with support for ondemand replication. d. Integrating Gospiffe library with gRPC to implement mTLS and secure interservice communication across distributed nodes. e. Using FastAPI, Docker containers and GPU-optimized data pipelines. f. Use of Python, PyTorch, and anomaly detection pipelines. Telecommuting/Remote Work is Permitted May telecommute Please copy and paste your resume in the email body (do not send attachments, we cannot open them) and email it to candidates at with reference in the subject line. Thank you.