Sarabjot Singh

AI Researcher

Bengaluru, Karnataka, India2 yrs 10 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Developed advanced GenAI applications for automotive industry.
Expert in building agentic AI systems for sales and engineering.
Proficient in multiple AI frameworks and cloud technologies.

Stackforce AI infers this person is a SaaS-focused AI Developer with expertise in advanced machine learning applications.

Contact

Skills

Core Skills

Artificial Intelligence (ai)PythonWeb DevelopmentReact.js

Other Skills

LangChainLlamaIndexAWS BedrockRAGLangGraphFastAPIChromaDBNeo4jFAISSOpenAI APIDockerClaude Sonnetdoclingmarker-pdfBack-End Web Development

About

Hi, I’m Sarabjot Singh, a AI Developer who spends his days wrangling LLMs, vector databases, and the occasional stubborn API. At Ashok Leyland, I build agentic and multi-agent AI systems that help engineers and sales teams work smarter — from OCR-powered document intelligence to RAG-driven assistants. My toolkit includes LangChain, LlamaIndex, LangGraph, AWS Bedrock, FastAPI, and React — basically, the Avengers of AI. When I’m not coding, I’m probably breaking my own projects just to rebuild them better. 😄

Experience

2 yrs 10 mos

Total Experience

1 yr 5 mos

Average Tenure

1 yr 10 mos

Current Experience

Ashok leyland

2 roles

AI Developer

Jul 2024 – Present · 1 yr 10 mos

Developed a suite of three GenAI applications for engineers and designers, enhancing document search and analytics.
Fine-tuned the LLaMA-3 8B model on proprietary automotive manuals, improving technical query resolution.
Created an agentic sales assistant that automated brochure delivery and product matching, significantly reducing manual lookup time.

PythonArtificial Intelligence (AI)

Artificial Intelligence Engineer

Jul 2024 – Present · 1 yr 10 mos

Delivered 3 GenAI applications using LangChain, LlamaIndex, AWS Bedrock, RAG for semantic search, analytics, and multi-modal querying.
Built agentic sales assistant with LangGraph and Claude Sonnet, automating brochure delivery, TCO calculation, and product matching—cutting lookup time by 70%.
Developed OCR pipelines with Python, docling, and marker-pdf to convert scanned PDFs into structured, queryable data for AI-powered Q&A.
Optimized RAG systems with ChromaDB, Neo4j, FAISS for precise retrieval of text and tabular data.
Stack: Python, LangChain, LangGraph, LlamaIndex, FastAPI, ChromaDB, Neo4j, FAISS, AWS (EC2, S3, Bedrock), OpenAI API, Claude Sonnet, Agentic AI, Docker.

LangChainLlamaIndexAWS BedrockRAGPythonLangGraph+7

Gdsc

Core team

Aug 2021 – Aug 2022 · 1 yr · Chandigarh, India

Led the design team at Google Developer Student Clubs (GDSC) in Chandigarh, focusing on user-centric design.
Conducted research on user interaction trends to create engaging graphics for community events.
Developed innovative strategies to enhance community engagement, resulting in a 30% increase in participation.

React.jsBack-End Web DevelopmentWeb Development