I am Sameer Vohra, an AI/ML Engineer specializing in production-grade LLM applications, Retrieval-Augmented Generation (RAG) pipelines, and intelligent agent systems. I design and ship end-to-end AI components—from hybrid retrieval and dense/sparse embeddings to cross-encoder reranking and Gemini-based agents with tool calling and memory management. I also focus on LLM evaluation, prompt engineering, and grounding verification to ensure accurate, policy-compliant outputs. I enjoy owning AI projects from architecture decisions through deployment, and I thrive in building reliable systems for real-world workflows such as HR and content platforms. I value clear collaboration, rigorous testing, and scalable, maintainable design.

Sameer Vohra

I am Sameer Vohra, an AI/ML Engineer specializing in production-grade LLM applications, Retrieval-Augmented Generation (RAG) pipelines, and intelligent agent systems. I design and ship end-to-end AI components—from hybrid retrieval and dense/sparse embeddings to cross-encoder reranking and Gemini-based agents with tool calling and memory management. I also focus on LLM evaluation, prompt engineering, and grounding verification to ensure accurate, policy-compliant outputs. I enjoy owning AI projects from architecture decisions through deployment, and I thrive in building reliable systems for real-world workflows such as HR and content platforms. I value clear collaboration, rigorous testing, and scalable, maintainable design.

Available to hire

I am Sameer Vohra, an AI/ML Engineer specializing in production-grade LLM applications, Retrieval-Augmented Generation (RAG) pipelines, and intelligent agent systems. I design and ship end-to-end AI components—from hybrid retrieval and dense/sparse embeddings to cross-encoder reranking and Gemini-based agents with tool calling and memory management. I also focus on LLM evaluation, prompt engineering, and grounding verification to ensure accurate, policy-compliant outputs.

I enjoy owning AI projects from architecture decisions through deployment, and I thrive in building reliable systems for real-world workflows such as HR and content platforms. I value clear collaboration, rigorous testing, and scalable, maintainable design.

See more

Experience Level

Expert
Expert
Expert
Expert
Expert
Expert
Intermediate
Intermediate
Intermediate
See more

Language

English
Fluent

Work Experience

AI/ML Engineer Intern at BSCIC Certifications Pvt. Ltd.
May 1, 2026 - Present
Designed and built a production RAG pipeline with Docling PDF ingestion, intelligent section-boundary chunking, dual-vector embeddings (dense all-MiniLM-L6-v2 + BM25 sparse) stored in Qdrant Cloud; implemented cross-encoder reranking (BAAI/bge-reranker-v2-m3) with a composite weighted score and sigmoid normalization; deployed a Gemini 2.5 Flash agent with a supervisor loop enabling precise tool execution, session memory via scratchpad, and long-term memory; built five tools for data queries, file uploads, and navigation; implemented a fault-tolerant async job queue with deduplication, retry logic, and full status lifecycle; integrated a FastAPI backend with a Django frontend to connect processing and UI; maintained two Qdrant collections for document embeddings and agent knowledge-base grounding.
AI Data Annotation Specialist (Freelance) at Innodata
March 1, 2026 - May 1, 2026
Evaluated 500+ LLM-generated outputs for correctness, grounding, reasoning validity, and policy compliance; performed grounding verification, multi-step reasoning validation, and reference-based response assessment following enterprise guidelines; developed working knowledge of LLM failure modes to inform production system design.
Digital Operations & Content Analyst at The Door Magazine
August 1, 2022 - June 1, 2025
Improved audience engagement by 35% through data-driven content strategy; coordinated 4–6 cross-functional teams on 50+ posts; built structured documentation and coordination workflows for content review and compliance.
QA Engineer / Game Tester at Babel Media (Keywords Studios)
October 1, 2021 - April 1, 2022
Filed 500+ reproducible bug reports in JIRA with severity tagging and clear reproduction steps; contributed to approximately 30% reduction in critical defects through structured regression testing.

Education

HND in Digital Technologies — AI Solutions & Applications at Global Business Studies (GBS Dubai)
January 1, 2025 - January 1, 2026
Bachelor of Computer Applications (BCA) — Computer Science at Guru Gobind Singh Indraprastha University
January 1, 2018 - September 1, 2022

Qualifications

Fundamentals of Large Language Models — Certified
January 11, 2030 - July 3, 2026
HND Semester 1 Distinction (Academic Distinction)
January 1, 2025 - December 31, 2025

Industry Experience

Software & Internet, Media & Entertainment, Professional Services