About Shaad Ahmad

Who Shaad Ahmad is, how engagements run, and what this GenAI consultancy does and does not take on.

Shaad Ahmad is a GenAI developer and LLM engineer based in Lucknow, India, working remotely with clients in India and Australia. Four years of Python, two of them spent delivering large language model applications that run in production rather than in a notebook.

The practice is deliberately a practice of one. Engagements are taken end to end, from the first scoping conversation through build, deployment and handover, which means the person who agrees the scope is the person who writes the code. Nothing is subcontracted and nothing is handed to a team that was not in the original conversation.

What the work looks like

Most engagements start with a corpus or a process rather than a model: a contract archive nobody can search, an inbox that turns into quotations by hand, a hiring funnel where every stage is a person copying between tabs. The interesting problem is almost never prompting. It is retrieval quality, schema validation, evaluation, and deciding which step a human still has to approve.

• Agentic RAG: Hybrid keyword and semantic retrieval, cross-encoder reranking and metadata filtering over a private knowledge base.
• Multi-agent orchestration: Stateful, tool-calling agents built with LangChain and LangGraph that plan, act and hold context across long-running workflows.
• LLM application backends: FastAPI and Next.js services with typed APIs, structured outputs, streaming and schema validation around the model layer.
• Document intelligence: OCR and layout-aware parsing that turns unstructured documents and email into clean, queryable data.
• Production hardening: Evaluation, tracing, guardrails, human-in-the-loop review, token budgets and cost control.

Selected engagements

• Legal Agentic RAG (Contracts Specialist Pty Ltd, Australia, 2025): Production RAG over 50,000+ contract and legal PDFs. Hybrid BM25 and dense retrieval with cross-encoder reranking, which measurably reduced the hallucination rate.
• AI hiring automation (Cartrends Auto Parts, India, 2025): A seven-stage pipeline from job posting through CV scoring, proctored assessment, async video interview, scheduling and offer letter to onboarding. Cut HR workload by roughly 95% per hire, leaving three human decision points.
• RFQ to quotation automation (Under NDA, metals trading, 2026): Email in, draft techno-commercial offer out. Gmail ingestion, structured extraction to a validated JSON schema, supplier matching against an embeddings knowledge base, and a review dashboard. Pre-filtering keeps non-RFQ mail away from the model to cut token spend, and nothing reaches a customer without human approval.

Technology

Python, TypeScript, FastAPI, Flask, Next.js, React, LangChain, LangGraph, OpenAI, Anthropic Claude, Google Gemini, Hugging Face, Pinecone, PostgreSQL, Docker, AWS S3, Tesseract OCR, pdfplumber.

Background

Graduated from Integral University, Lucknow. Based in Lucknow, India and available for remote engagements worldwide.