AI Platform EngineerJan 2024 – Present
AI & Backend Systems
Omni RPA Inc — San Jose↳ Agentic Solutions Pvt Ltd, Hyderabad
- MCP server and client from scratchBuilt the entire MCP implementation from scratch. Tool gating with prerequisite chains, dynamic schema conversion, attribute matrix filtering, value validation against live session state.
- Knowledge graph and RAG infraOwn the entire knowledge graph stack. Neo4j to FalkorDB migration, hybrid RAG on Weaviate with gRPC, two-layer semantic cache (Redis + RedisVL), event-driven ingestion pipeline on NATS.
- DAG-based multi-agent orchestrator8 agent types, topological sort with cycle detection, parallel execution, dependency injection, persistence, WebSocket streaming.
- Semantic memory on PostgreSQL with pgvectorEmbedding-based retrieval, rolling summarization, optimistic concurrency control. Runs in background, enables continuity across sessions.
- Performance engineering across servicesDatabase indexing, bulk operations, L1/L2 cache layering, parallel graph calls, redundant query elimination.
- Own 3-5 repos end-to-end, primary on-callManage all PRs/issues, containerize services, ship to production. 2hr worst-case incident recovery.