AK Ameet Kumar
Open to opportunities in ML Engineering

Machine Learning
Engineer.

I build intelligent products that move from model experimentation to dependable, application-facing systems.

Get in touch ↗ LinkedIn ↗
📍 Milano, Italy 3+ years experience AI / ML / GenAI
ML ENGINEERING01
01focus = [
02"recommendation",
03"NLP", "vision",
04"generative-ai"
05]
06
07ship(models, APIs, infra)
08✓ production-ready
40%less manual review
30%screening time saved
30%engagement lift
01 — About

Building ML systems,
not just models.

Machine Learning Engineer with 3+ years of experience developing and deploying intelligent software systems across recommendation, personalisation, NLP, computer vision, and Generative AI.

Experienced in taking machine-learning models from experimentation into application-facing services, with hands-on work spanning data preparation, model integration, inference optimisation, API development, containerisation, automated validation, and cloud deployment.

Strong interest in dependable ML platforms, recommendation infrastructure, and high-performance inference systems, with a focus on maintainable code, measurable experimentation, and clear technical communication.

02 — Experience

Experience that ships
into production.

From applied AI workflows to model serving, testing, and deployment.

EMPIREONE

AI/ML Engineer

May 2024 — Present
500+calls processed / day
40%manual review effort reduced
30%screening time reduced
  • Designed an automated quality-assurance platform for call-centre recordings using speech recognition and language models, processing approximately 500 calls per day and reducing manual review effort by around 40%.
  • Developed an AI recruitment workflow covering candidate document processing, profile scoring, ranking, and automated communication, reducing average screening time by approximately 30%.
  • Developed backend services for serving machine-learning models through REST interfaces consumed by multiple application components.
  • Built model evaluation workflows for comparing alternative models, analysing failure cases, and measuring changes in prediction quality and runtime behaviour.
  • Packaged inference services with Docker and introduced automated build and verification workflows through Git-based CI pipelines.
  • Prepared technical reports covering architecture decisions, experiment results, system limitations, and recommended improvements.
  • Supported junior engineers with ML deployment, model evaluation, retrieval-based systems, and production-oriented development practices.
  • Investigated production defects and proposed changes to data processing, model behaviour, and service architecture based on observed system performance.
Appedology

Junior AI Engineer

Sep 2023 — May 2024
50%faster ad-hoc SQL querying
2w → 4dprototype-to-demo cycle
30%user engagement increase
  • Designed and deployed Multi-PDF Chatbot and Text-to-SQL Chatbot solutions using open-source LLMs (LLaMA 2, Mistral), LangChain, and FAISS; reduced ad-hoc SQL query time for non-technical staff by ~50%.
  • Added unit, integration, and end-to-end test coverage for prediction behaviour and API-level failures.
  • Created technical notes and evaluation reports explaining system behaviour, experiment outcomes, limitations, and proposed improvements.
  • Led an R&D initiative to build a multi-agent SDLC automation framework, coordinating specialised agents for planning, coding, and review — reducing prototype-to-demo cycle from 2 weeks to 4 days.
  • Developed a collaborative-filtering + NLP recommendation engine that increased user engagement by 30% in A/B testing.
  • Implemented a deep-learning segmentation model (U-Net) for medical image analysis in partnership with healthcare professionals, achieving 88% IoU on validation data.
  • Stayed current with AI research; introduced weekly paper-reading sessions and standardised evaluation protocols across the team.
03 — Technical Stack

A toolkit for
production AI.

I work across the full path from data and experimentation to model serving, APIs, infrastructure, and validation.

Machine Learning & Recommendation

PythonPyTorchTensorFlow / KerasScikit-learnRecommendation SystemsCollaborative FilteringRankingFeature EngineeringModel EvaluationA/B TestingDeep LearningNLPComputer Vision

ML Engineering & APIs

FastAPIREST APIsModel ServingInference PipelinesDockerKubernetesCI/CDMLflowGitGitHub

Data & Processing

SQLPandasNumPyPySparkMySQLPostgreSQLMongoDBFAISSData PipelinesFeature Processing

Cloud & Infrastructure

AWS SageMakerAWS S3AWS LambdaAWS EC2Azure MLGoogle Vertex AI

Generative AI

OpenAI APIMistralLLaMAHugging FaceLangChainLlamaIndexPEFTRAGAgentic AI

Quality & Engineering

Automated TestingModel ValidationBenchmarkingError AnalysisLoggingMonitoringTechnical DocumentationAPI Documentation
04 — Education
2024 — 2026
MSc · Expected September 2026

Artificial Intelligence for Science and Technology

Università degli Studi di Milano-Bicocca · Milano, Italy

Deep Learning · NLP · Computer Vision · Responsible AI · Statistical Learning
2019 — 2023
Bachelors

Bachelor of Software Engineering

Sukkur IBA University · Sukkur, Pakistan

Strong focus on algorithms, data structures, and software architecture
Training
Coursera

Applied Machine Learning in Python

05 — Languages
EnglishC1Full professional proficiency
ItalianA2Elementary · actively studying
06 — Contact

Let's build something
intelligent.

Open to Machine Learning, AI Engineering, Recommendation, and Generative AI opportunities in Italy and internationally.

Emailameet4562@gmail.com ↗ LinkedInlinkedin.com/in/ameet-kumar-487975202 ↗
Location20162 Milano, Italy