Anmol S.

AI Engineer focusing on AI Agents, LLMs, RAGs, and n8n automations. I develop production-level AI solutions with OpenAI, Claude, LangGraph, CrewAI, MCP, Python, FastAPI, vector database, and contemporary automation technology - not just POCs or demo-day chatbots. I assist startups and corporations to take their AI from the concept stage to actual production systems - whether it is AI Agents, intelligent processes, RAG use cases, voice AI, document processing, data flows, or AI-powered automation for business purposes. Enterprise and Compliance Experience: I have 20+ years of experience in software/data engineering and have worked on enterprise-level projects at DXC, Lincoln National, Jackson National, and Cisco. My experience includes the development of systems for environments where SOC 2, HIPAA, GDPR, and CMMI Level 5 requirements applied. Such experience is important in case an AI system requires not only working prototype but also works in a complex environment such as finance, insurance, healthcare, etc. What I build AI Agents & Agentic Systems • Multi-agent systems using LangGraph, CrewAI, AutoGen and MCP • Tool agents integrated with various APIs, databases and business applications • Agent orchestration, agent memory, agent routing and human-in-the-loop workflows • Research, data extraction, reporting and operational workflow automations LLM & RAG Applications • Enterprise production RAG solutions • Semantic/hybrid search using Pinecone, Weaviate, Qdrant, Milvus and pgvector • LangChain, LlamaIndex and GraphRAG applications • Document ingestion, chunking, embedding, retrieval and evaluation • OpenAI, Anthropic Claude, Gemini and open-source LLM integrations • Structured LLM prompts and function calls AI Automation • n8n-based AI-powered workflows and business processes automation • Integration of AI with CRMs, databases, various APIs and internal applications • Zapier, Make and custom Python-based automations • Automation pipelines from web/data extraction to processing, analysis, reporting • Sales, support, operations, research and document processing workflows powered by AI Voice AI • AI voice agents and conversational workflow automations • Vapi, Retell, Twilio, Deepgram, ElevenLabs and other voice technologies • Automated intake, qualification, scheduling, support and follow-up workflows The Production Engineering Behind the AI : One of my strongest advantages is that I have a solid software and data engineering background, which means I don't see AI as an isolated API integration problem. I develop the infrastructure needed for AI to work reliably: • Python, FastAPI, Node.js and REST/GraphQL APIs • PostgreSQL, MongoDB, Redis and other databases • AWS, Azure and Google Cloud • Docker, Kubernetes and Terraform • CI/CD and production deployment • Logging, monitoring and error handling • Data validation and pipeline reliability • MLOps and model monitoring Data Engineering for AI : A lot of AI projects are doomed to failure due to the underlying data being of low quality. I create the data layer AI projects depend on: • ETL/ELT pipelines • Airflow and dbt • Spark and Python data processing • Snowflake, BigQuery and Databricks • Kafka/Kinesis for streaming workloads • Processing of structured and unstructured data • Data extraction, transformation and enrichment pipelines If you are struggling with your AI project delivering bad results due to the underlying data being messy, partial or hard to get, this is where I can help the most. My AI Stack: - AI/LLM: OpenAI, Claude, Gemini, Hugging Face, Amazon Bedrock - AI Agents: LangGraph, CrewAI, AutoGen, MCP, Semantic Kernel - RAG: LangChain, LlamaIndex, GraphRAG, Pinecone, Weaviate, Qdrant, Milvus, pgvector, FAISS - Automation: n8n, Make, Zapier, Flowise, LangFlow - Voice AI: Vapi, Retell, Twilio, Deepgram, ElevenLabs, Whisper - LLM Engineering: Prompt Engineering, Function Calling, Structured Outputs, Fine-tuning, LoRA/QLoRA, PEFT, vLLM - Backend: Python, FastAPI, Django, Flask, Node.js, Express.js - Cloud/Infrastructure: AWS, Azure, GCP, Docker, Kubernetes, Terraform - Data: Snowflake, BigQuery, Databricks, Airflow, dbt My approach is to build for production and transfer, and not just for a good demo – with clear architecture, documentation, monitoring, and maintainable code which can be further worked on by your internal team. In case you’re looking for an AI Engineer who can help you build AI Agents, LLMs/RAGs, n8n automations, voice AI or the underlying infrastructure of all of the above, do get in touch with me. -- Best Regards, Anmol