Rohit Goswami profile photo

Hi, I'm

Rohit Goswami

AI Engineer

ML/DL/NLP | RL | Multi-Agent Systems | Security | RAG | A2A | MCP | Fine-Tuning

Get To Know More

About Me

With 8+ years of experience, I specialize in designing and shipping enterprise AI platforms that are secure, observable, and scalable. My work spans multi-agent orchestration, LLM security guardrails, and production-grade MLOps on Kubernetes.

Multi-Agent Orchestration

Supervisor/Hierarchical patterns, A2A, tool-calling with streaming hooks for complex agentic workflows.

LLM Security

Scanner pipelines, PII redaction, policy enforcement, and continuous monitoring for safe AI deployments.

MLOps on Kubernetes

Helm charts, Prometheus/Grafana observability, GPU monitoring, and autoscaling for reliable model serving.

Tooling & Workflows

FastAPI services, Temporal orchestration, Redis caching, and OpenAI Agents SDK for production automation.

Browse My Recent

Projects

Hierarchical Multi-Agent Orchestrator

Hierarchical Multi-Agent Orchestrator

  • Supervisor–child task delegation
  • Tool routing & structured outputs
  • Planning vs execution separation
PythonLangGraphFastAPIAgents
Google A2A Interoperability Demo

Google A2A Interoperability Demo

  • Google A2A protocol in action
  • Agent Cards for discovery & negotiation
  • Budget-constrained orchestration
PythonA2A ProtocolAgentsInteroperability
Temporal.io Orchestrated RAG Application

Temporal.io Orchestrated RAG Application

  • Temporal.io durable workflow orchestration
  • Multiple chunking strategies & scalable ingestion
  • Interactive querying via Streamlit interface
PythonTemporal.ioLangChainRAGVector DatabaseStreamlitLLM
OpenAI Realtime Voice Assistant

OpenAI Realtime Voice Assistant

  • Live voice conversations via OpenAI Realtime API
  • Low-latency streaming responses
  • Interactive speech interface with Streamlit UI
PythonOpenAI Realtime APIStreamlitWebRTC
LLM Distributed Finetuning

LLM Distributed Finetuning

  • LoRA / QLoRA + DeepSpeed ZeRO-3 training
  • Scalable multi-GPU workflows via LLaMA-Factory
  • Production configs, Docker, and FastAPI serving
PythonDeepSpeedLoRAQLoRALLaMA-FactoryDockerFastAPI
URL Shortener in Go

URL Shortener in Go

  • Base62 ID generation with clean REST API
  • Dockerized with full test coverage
  • Minimalist UI with redirect analytics
GoGinBase62DockerTests

My Academic Background

Education

New Jersey Institute of Technology

Master of Science - MS, Data Science

NJIT M.S. in Data Science Program

Dr. Babasaheb Ambedkar Marathwada University (BAMU), Aurangabad

Bachelor's degree, Electrical and Electronics Engineering

Get in Touch

Contact Me

Have a question or want to collaborate? I'd love to hear from you. Drop a message and I'll get back to you as soon as I can.