
Generative AI Developer | Agentic AI Developer | AI Engineer | Forward Deployed Engineer (FDE) | AI Solutions Engineer
Self-Employed · Aug 2025 – Present
- Hands-on GenAI developer with experience building LLM applications, RAG pipelines, AI agents, and Agentic AI workflows.
- Built GenAI services with Python and FastAPI, exposing AI capabilities through REST APIs and application backends.
- Hands-on work with RAG, semantic retrieval, embeddings, vector databases, reranking, metadata filtering, guardrails, LLM evaluation, prompt engineering, and Agentic AI loops.
- Worked with Deep Agents, Model Context Protocol (MCP Server), Anthropic API (Claude), and LLM fine-tuning fundamentals.
- Built RecruitOS AI, an end-to-end RAG + Agentic AI recruitment platform, combining GenAI engineering with 12+ years of recruiting-domain knowledge.
- Strong understanding of real-world recruiting workflows, used as domain expertise when designing AI-based resume screening, candidate retrieval, matching, and ranking workflows.
- Hands-on experience in developing GenAI, RAG, and Agentic AI solutions, with a focus on building practical, production-oriented AI applications.
- Built a RAG pipeline from document ingestion and text extraction through chunking, embeddings, vector storage, similarity search, retrieval, and LLM response generation.
- Improved retrieval quality using embeddings, vector databases, Top-K retrieval, reranking, and metadata filtering.
- Worked on LLM evaluation, hallucination reduction, guardrails, and response validation for more reliable outputs.
- Built AI agent workflows that plan tasks, reason through steps, call tools, observe results, and continue through multi-step Agentic AI loops.
- Worked with Deep Agents for workflows requiring planning, memory, and autonomous execution.
- Integrated MCP Server concepts so agents can communicate with external tools, services, and data sources.
- Worked with AI agents using the Anthropic API (Claude), including task planning, tool calls, and multi-step execution.
- Applied structured prompts, role-based prompting, and context management to improve consistency of LLM outputs.
- Hands-on with LLM fine-tuning fundamentals and adapting foundation models for specific use cases.
- Built GenAI services in Python/FastAPI and connected them to application backends through REST APIs.
- Used Git/GitHub for source control and project development.





