Aspiring AI / Machine Learning Engineer | Software Development

Computer Science graduate from the University of Sussex with a strong foundation in software development, Artificial Intelligence, and Machine Learning. Currently a Junior Software DevOps & Cloud Engineer intern at MetLife UK, working on agentic AI solutions. Experienced in applied ML projects including NLP-based automation tools, computer vision pipelines, and CNN models. Advanced in Python, C, and Java, with exposure to building secure applications and scalable systems.

I enjoy building intelligent systems that turn ideas into working products, whether that's experimenting with machine learning models, improving application performance, or refining system design. I'm especially interested in how AI can be applied responsibly and creatively, and I like exploring new tools through hands-on projects. When I'm not coding, I enjoy learning about emerging technologies, AI ethics, and experimenting with side projects that push my technical boundaries.

Sahil Shindgikar
Based in Brighton, England
sahilshindgikar@gmail.com
View Résumé
GitHubLinkedIn
Scroll, the work speaks from here
Machine LearningComputer VisionNLPCloud SystemsSoftware EngineeringGPU ComputingMachine LearningComputer VisionNLPCloud SystemsSoftware EngineeringGPU ComputingMachine LearningComputer VisionNLPCloud SystemsSoftware EngineeringGPU Computing

01Capabilities

The toolbox

Languages, frameworks, and infrastructure I reach for, grouped by the problems they solve.

Programming Languages

PythonJavaCTypeScriptHTML/CSSJavaScript

AI/ML

Machine LearningComputer VisionDeep Learning FrameworksLogistic RegressionSupervised LearningUnsupervised LearningVisulization Techniques

Cloud Computing

AWSEC2Cloud DeploymentCloud InfrastructureCloud StorageIAM Basics

Parallel Programming

CUDAGPU ComputingParallel AlgorithmsThreadingSynchronizationNVIDIA architecture

Database

MySQLDynamoDB

Tools

GitGitHubDockerWSLHeidiSQLClaude CodeGitHub Copilot

02Experience

Where the work happened

-

Brighton, UK (Hybrid)

Junior Software DevOps & Cloud Engineer

MetLife UK

Building an agentic AI system on Microsoft Copilot Studio that helps employees discover key internal documents and get support within company guidelines, deployed across Microsoft Teams and other integrations.

Microsoft Copilot StudioMicrosoft TeamsAgentic AI

-

Brighton, UK

Committee Member

HackSussex

Designing merchandise and capturing event photography to drive engagement for the Annual Hackathon and the student developer community.

  • Graphic Design: Co-lead the creation of digital and print assets for high-profile events like the Annual Hackathon, ensuring all promotional material aligns with the society's branding.
  • Photography: Serving as an official photographer for society events, capturing high-quality images of weekly workshops, socials, and large-scale competitions for marketing and social media use.

-

London, United Kingdom

Work Experience Placement

Just Group Plc

Gained valuable exposure to the financial services sector, developing a foundational understanding of software development life cycles and Agile methodologies. Explored the intersection of technology and business through AI automation, cloud architecture, and human behavioral science. Proven ability to collaborate effectively within a team-oriented corporate environment.

Microsoft copilotGoogle VeoRapid prototypingMicrosoft Azure 365

-

Brighton, UK

Team Leader

Engineers Without Borders UK (Student Chapter)

Led a successful university project team for the Global Design Challenge 2024, designing an independent solar system to address electricity shortages in Pu Ngaol, Cambodia.

  • Secured a win in the respective hub by leading a student team in designing a sustainable off-grid solar solution
  • Orchestrated strategic planning and workflow management, successfully re-allocating tasks during a team shortage to meet strict academic deadlines
  • Developed a low-maintenance energy proposal that addressed the specific economic and environmental needs of the Pu Ngaol village

03Selected Work

Things I've built & shipped

An index of the projects that taught me the most, machine learning systems, distributed platforms, and the pipelines between them.

01

Resume Genie

Problem
Manual job searching is slow, and resumes tuned by hand routinely fail ATS keyword filters.
System
Python + Generative AI pipeline that parses job descriptions, programmatically optimizes resumes for ATS compatibility, and generates tailored interview prep materials.
Result
Turns the job search into an automated, data-driven process - discovery through tailored application.
Stack
PythonGenerative AIREST APIs
Code ↗
02

Event-Driven Media Platform

Problem
Uploading, transcoding, and streaming video at scale without blocking user-facing services.
System
Spring Boot microservices coordinated by Kafka workers: JWT-authenticated async uploads, FFmpeg multi-resolution transcoding, HLS adaptive streaming, PostgreSQL metadata, MinIO object storage.
Result
A working distributed pipeline demonstrating practical microservices design and event-driven coordination end to end.
Stack
Spring BootKafkaFFmpegPostgreSQLMinIODockerJWT
Code ↗
03

Machine Learning Face Alignment

Problem
Detecting facial keypoints precisely across varied face images - where classical ML meets deep learning.
System
Custom CNN combined with Ridge Regression, built in PyTorch and OpenCV, predicting 15 facial landmarks per image.
Result
A robust detection pipeline with precise feature mapping and real-time coordinate visualization.
Stack
NumPyOpenCVPyTorchMatplotlib
Code ↗
04

Fashion-MNIST GAN + CNN

Problem
Classify clothing images - then go further and generate convincing new ones from random noise.
System
PyTorch CNN classifier with Batch Normalisation and Adam optimisation, plus an unconditional GAN extended to a Conditional GAN generating specific clothing categories on demand.
Result
92% CNN test accuracy; Conditional GAN reached 78% label accuracy across all 10 classes.
Stack
PythonPyTorchGANCNNFashion-MNISTMatplotlib
Code ↗
05

Propaganda Detection NLP

Problem
Detecting propaganda techniques in news articles - both classifying known spans and finding the spans themselves.
System
TF-IDF and MiniLM baselines benchmarked against fine-tuned DeBERTa-v3-large, plus a novel two-stage span detector pairing DeBERTa token classification with a separate technique classifier.
Result
Full comparison of classical vs. transformer approaches, evaluated on macro F1, token-level span F1, and joint exact match.
Stack
PythonTransformersHugging FaceNLP
Code ↗
Full archive on GitHub ↗

04Education

The academic record

-

Brighton, England

Bachelor of Science, Computer Science

University of Sussex

First Class Honors

  • 2024 Global Design Challenge hub winners: First in our hub in the university GDC 2024.

-

Bangalore, India

Bachelor of Technology, Computer Science

PES University

Further Study - Certifications

05Beyond the Code

The rest of the picture

How I work, what I'm chasing right now, and what fills the hours when the laptop is closed.

Working Principles

  1. Continuous Learning

    Actively strengthening fundamentals while staying current with emerging AI and software practices.

  2. Practical Impact

    Building solutions that solve real-world problems rather than purely experimental prototypes.

  3. Responsible AI

    Thinking critically about ethical implications, fairness, and long-term impact of intelligent systems.

  4. Collaboration & Feedback

    Valuing peer review, open discussion, and iterative improvement.

Currently Exploring

  • Applied machine learning pipelines and experimentation workflows
  • Model optimization, evaluation, and deployment strategies
  • Computer vision and representation learning
  • Cloud-native architectures for scalable ML systems
  • Open-source contribution and collaborative development

Off the Clock

Building personal coding projectsGym and fitness trainingHiking and outdoor activitiesPhotographyPlaying guitarCooking

Languages Spoken

  • EnglishFluent
  • FrenchIntermediate
  • HindiNative
  • MarathiNative

06Contact

Let’s build something worth shipping.

I’m always interested in hearing about new opportunities, projects, or just having a chat about technology and AI.

Availability

Interning at MetLife

Currently in

Brighton, England