Mansi Rathod profile photo 1 on Tal

Mansi Rathod

ML & AI Engineer | Production LLM Systems | LLM Inference Optimization

Greater Bengaluru Area

6+ years of work experience

Apra Labs.
Jan 20215 yrs 8 mosPresent

Bengaluru, Karnataka, India

AI on edge devices, Fine-tuning LLMs, Production LLM deployment, Model optimizations, Accelerated Inferencing using NVIDIA, Speech Recognition, Video Analytics, Natural Language Processing, C++ and .net developer
IIIT-Naya Raipur
Jan 20201 yr 1 moJan 2021

Greater Raipur Area

E-Cell, IIIT-Naya Raipur
20182 yrs 1 moJan 2020

Greater Raipur Area

Mansi’s AI Stack

claude code
claude code
hugging face
hugging face
langchain
langchain
mcp
mcp
openclaw
openclaw
llama.cpp
llama.cpp

Education

Bachelor of Technology, Electronics and Communications Engineering

International Institute of Information Technology,naya raipur

20174 yrs2021

Join Mansi and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Mansi Rathod on Tal

Headline

ML & AI Engineer | Production LLM Systems | LLM Inference Optimization

About

I’m a Machine Learning Engineer who enjoys turning ambitious AI ideas into production systems that actually scale. Over the past 5+ years, I’ve worked across computer vision, speech AI, and large language models , building end-to-end systems and taking them from research to reliable, 24/7 deployments. What I work on most: • LLM fine-tuning • RAG systems and vector search pipelines • LLM inference optimization (TensorRT, quantization, Triton) • Multimodal AI (vision + speech + text) • Edge-to-cloud AI deployments Recently, I optimized a 32B parameter LLM on A100 80GB by building a native INT4 TensorRT engine — reducing memory usage significantly while enabling concurrent high-throughput inference. I enjoy working at this intersection of models + systems + performance. Beyond LLMs, I’ve: – Built large-scale video analytics pipelines across thousands of camera feeds – Deployed real-time speech-to-text systems – Designed multimodal search and Q&A systems with citation-backed retrieval I like solving practical AI problems, especially where performance, scalability, and cost efficiency matter. Currently exploring opportunities in production LLM systems, AI infrastructure, and multimodal GenAI products.

Location

Greater Bengaluru Area

AI stack

claude code

hugging face

langchain

mcp

openclaw

llama.cpp

Work experience

Machine Learning Engineer at Apra Labs., Jan 2021 - Present · 5 yrs 8 mos, Bengaluru, Karnataka, India, AI on edge devices, Fine-tuning LLMs, Production LLM deployment, Model optimizations, Accelerated Inferencing using NVIDIA, Speech Recognition, Video Analytics, Natural Language Processing, C++ and .net developer

Training and Placement cell Head Coordinator at IIIT-Naya Raipur, Jan 2020 - Jan 2021 · 1 yr 1 mo, Greater Raipur Area

Head of Documentation department at E-Cell, IIIT-Naya Raipur, 2018 - Jan 2020 · 2 yrs 1 mo, Greater Raipur Area

Education

Bachelor of Technology, Electronics and Communications Engineering, International Institute of Information Technology,naya raipur