Anupam Misra profile photo 1 on Tal

Anupam Misra

Senior Machine Learning Engineer - Agentic AI & MLOps at scale

Bengaluru

7+ years of work experience

Eli Lilly and Company
Oct 20232 y 10 moPresent

Bengaluru

Enterprise Agentic AI Platform: * Built and scaled a microservices-based agentic AI platform for the commercial org, dramatically cutting agent time-to-production through ready-made templates and a managed orchestration, runtime, and observability layer. * Designed for high-concurrency use with async distributed execution * Centralized logging/tracing across the full stack, and cost-optimized LLM routing. Tools: FastAPI, LangGraph, Kubernetes, Celery, RabbitMQ, Redis, Prometheus, Grafana, KEDA, Docker, LLM Generative AI Products: Delivered multiple production AI tools across the org: * Self-serve natural-language-to-SQL analytics assistant for cross-brand reporting * MCP-based marketing-mix-modeling tool used to evaluate channel spend impact. Tools: LangGraph, CrewAI, Databricks, Qdrant, MCP Multi-Touch Attribution Modelling: * Designed and trained deep learning attribution models for large-scale consumer-level marketing attribution * Used distributed hyperparameter tuning on Kubernetes to significantly reduce training time. * Provided parallelized batch based inference for all target customers. Tools: PyTorch, Optuna, AWS, MLflow
PwC India
Apr 20236 mosSep 2023

Bengaluru

Preventing equipment failure at Shell across 20+ operating sites. * Developed end-to-end MLOps pipelines for outlier detection on streaming sensor data across 470+ assets in 20+ operating sites for Shell * Implemented data drift detection via PCA reconstruction error to close the observability loop, enabling pre-emptive retraining of degraded models * Applied a kernel derivative method to extract the top two sensor tags in a One-Class SVM model with 75% more accuracy than previous Z-score based method Firm development initiatives: • Fine-tuned abstractive dialogue summariser using Pytorch and Google-T5 in Azure ML Studio. The real-time endpoint API was consumed in a web application. The web application used Google’s STT API to convert audio dialogue to text. • Onboarded 13 interns, associates and senior associates into the team.
PwC India
Nov 20211 yr 5 mosMar 2023

Bengaluru, Karnataka, India

Designed and developed a full stack web application for HR analytics: • Guided four colleagues to build the application using Python, Flask, HTML, JS and CSS. The application was deployed on Microsoft Azure App Services with VS Code easy-deploy. • Attrition prediction was done via a logistic regression model with F1 score of 0.91. The application was designed to accept data and display attrition probability with local importance using shapely values. • Candidates were recommended through fuzzy matching resumes and uploaded job description.
Reliance Industries Limited
Jul 20191 yr 7 mosJan 2021

Jamnagar Area, India

1. Built interactive dashboards for key stakeholders 2. Ensured reliability of 130 MW of electrical equipment through coordination with other departments. 3. Led a team of thirteen men for executing daily tasks. 4. PR/PO management and work order planning through SAP.
Reliance Industries Limited
Jul 20181 yr 1 moJul 2019

Jamnagar, Gujarat, India

Underwent plant operation, electric equipment and safety related training in DTA Pet Coke Gasfication project.
GitHub
@misra-anupam

Anupam’s AI Stack

claude code
claude code
langchain
langchain
langgraph
langgraph
mcp
mcp
OpenRouter
OpenRouter
Qdrant
Qdrant

Education

Praxis Business School

PGP Data Science, Data science

Praxis Business School

20212021
Manipal institute of technology

Bachelor of Technology (B.Tech.), Electrical and Electronics Engineering

Manipal institute of technology

20144 yrs2018

Join Anupam and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Anupam Misra on Tal

Headline

Senior Machine Learning Engineer - Agentic AI & MLOps at scale

About

Senior Machine Learning Engineer with 7+ years of experience building production-grade Agentic AI systems and large-scale MLOps pipelines — from architecting a distributed microservices-based multi-tenant LLM platform to building distributed ML pipelines for 20M consumers and managing model lifecycles for 470+ assets across 20+ industrial sites. Tools and technologies - Python, LangGraph, CrewAI, FastAPI, Redis, RabbitMQ, Celery, Flower, Docker, Kubernetes, Loki, Promtail, Grafana, Prometheus, JS, AlibiDetect, Prefect, Pytorch, AWS, Azure. Areas of interest - Agentic systems and platforms, MLOps at scale.

Location

Bengaluru

AI stack

claude code

langchain

langgraph

mcp

OpenRouter

Qdrant

Work experience

Senior Machine Learning Engineer at Eli Lilly and Company, Oct 2023 - Present · 2 y 10 mo, Bengaluru, Enterprise Agentic AI Platform: * Built and scaled a microservices-based agentic AI platform for the commercial org, dramatically cutting agent time-to-production through ready-made templates and a managed orchestration, runtime, and observability layer. * Designed for high-concurrency use with async distributed execution * Centralized logging/tracing across the full stack, and cost-optimized LLM routing. Tools: FastAPI, LangGraph, Kubernetes, Celery, RabbitMQ, Redis, Prometheus, Grafana, KEDA, Docker, LLM Generative AI Products: Delivered multiple production AI tools across the org: * Self-serve natural-language-to-SQL analytics assistant for cross-brand reporting * MCP-based marketing-mix-modeling tool used to evaluate channel spend impact. Tools: LangGraph, CrewAI, Databricks, Qdrant, MCP Multi-Touch Attribution Modelling: * Designed and trained deep learning attribution models for large-scale consumer-level marketing attribution * Used distributed hyperparameter tuning on Kubernetes to significantly reduce training time. * Provided parallelized batch based inference for all target customers. Tools: PyTorch, Optuna, AWS, MLflow

Senior Associate at PwC India, Apr 2023 - Sep 2023 · 6 mos, Bengaluru, Preventing equipment failure at Shell across 20+ operating sites. * Developed end-to-end MLOps pipelines for outlier detection on streaming sensor data across 470+ assets in 20+ operating sites for Shell * Implemented data drift detection via PCA reconstruction error to close the observability loop, enabling pre-emptive retraining of degraded models * Applied a kernel derivative method to extract the top two sensor tags in a One-Class SVM model with 75% more accuracy than previous Z-score based method Firm development initiatives: • Fine-tuned abstractive dialogue summariser using Pytorch and Google-T5 in Azure ML Studio. The real-time endpoint API was consumed in a web application. The web application used Google’s STT API to convert audio dialogue to text. • Onboarded 13 interns, associates and senior associates into the team.

Associate at PwC India, Nov 2021 - Mar 2023 · 1 yr 5 mos, Bengaluru, Karnataka, India, Designed and developed a full stack web application for HR analytics: • Guided four colleagues to build the application using Python, Flask, HTML, JS and CSS. The application was deployed on Microsoft Azure App Services with VS Code easy-deploy. • Attrition prediction was done via a logistic regression model with F1 score of 0.91. The application was designed to accept data and display attrition probability with local importance using shapely values. • Candidates were recommended through fuzzy matching resumes and uploaded job description.

Manager at Reliance Industries Limited, Jul 2019 - Jan 2021 · 1 yr 7 mos, Jamnagar Area, India, 1. Built interactive dashboards for key stakeholders 2. Ensured reliability of 130 MW of electrical equipment through coordination with other departments. 3. Led a team of thirteen men for executing daily tasks. 4. PR/PO management and work order planning through SAP.

Graduate Engineering Trainee at Reliance Industries Limited, Jul 2018 - Jul 2019 · 1 yr 1 mo, Jamnagar, Gujarat, India, Underwent plant operation, electric equipment and safety related training in DTA Pet Coke Gasfication project.

Education

PGP Data Science, Data science, Praxis Business School

Bachelor of Technology (B.Tech.), Electrical and Electronics Engineering, Manipal institute of technology

Public links

github: https://github.com/misra-anupam