Software engineer focused on backend systems, distributed architectures, and the path from first commit to production.
I'm a Master's student in Software Engineering at the University of Tartu, building the kind of systems most teams need but few engineers fully own end-to-end. My work centers on Java and Spring Boot microservices, event-driven communication over message brokers, secure REST APIs, and the delivery pipelines that get all of it into production.
I treat software engineering as a craft: clean service boundaries, well-modeled domains, observable systems, and code that future maintainers can read without flinching. On the full-stack side I'm comfortable taking a feature from PostgreSQL schema to Vue or Next.js UI; on the AI / ML side I've trained and benchmarked supervised classifiers for production fraud detection.
I bring a product engineering mindset — I built and shipped Üüriturg, a rental aggregator now serving the Tartu market live, after experiencing the problem it solves first-hand. Shipping isn't a phase of the project for me; it's the proof.
Software Engineering Internships • Backend • DevOps • Cloud / Platform • Security Engineering • Tartu / Tallinn / Remote EU
| Domain | Proficiency | Details |
|---|---|---|
| Supervised Classification | ⬛⬛⬛⬛⬜ | Logistic Regression, Linear SVM, Random Forest, XGBoost benchmarking |
| Natural Language Processing | ⬛⬛⬛⬛⬜ | TF-IDF vectorization, tokenization, stopword removal, n-gram features |
| Model Evaluation | ⬛⬛⬛⬛⬜ | Precision / Recall / F1 trade-offs, cross-validation, confusion analysis |
| Feature Engineering | ⬛⬛⬛⬜⬜ | 10,000-feature TF-IDF pipelines, categorical encoding, text normalization |
| Production ML Pipelines | ⬛⬛⬛⬜⬜ | scikit-learn + pandas pipelines, reproducible training, model serialization |
| Frameworks & Libraries | ⬛⬛⬛⬛⬜ | scikit-learn, XGBoost, NLTK, NumPy, pandas |
🏆 Üüriturg — Rental Market Aggregator (UT ICS Student Project Contest · Live)
A microservices-based rental market aggregator for Tartu — every listing from five Estonian property portals, unified in one free, centralized discovery platform. Presented at the University of Tartu ICS Student Project Contest poster session and currently serving the live market.
| Attribute | Detail |
|---|---|
| Stack | Java 21 · Spring Boot · Spring Cloud Gateway · Resilience4j · RabbitMQ · PostgreSQL · Vue 3 · Leaflet · Chart.js · Docker · nginx |
| Scale | 4 microservices · 600+ listings · 5 source portals · 6-hour ingestion cycle |
| Performance | Event-driven price alerts delivered to subscribers in ~30 seconds end-to-end |
| Security | Per-service auth boundaries · input validation · GDPR-aware data handling roadmap |
| Impact | Live on university infrastructure · presented at UT ICS Student Project Contest |
| Repository | github.com/hashimminhas/Rental-Market |
Architected with strict service boundaries: a Scraper Service normalizes heterogeneous listing data, RabbitMQ carries listing.matched events to an Alert Service that fans out subscriber notifications, and an Analytics Service pre-computes neighborhood price snapshots to keep dashboards responsive under load. Deployed via Docker Compose behind an nginx reverse proxy on university HPC infrastructure.
🎓 UniCampus — University Campus Management Platform
An 8-microservice university campus management system designed with Domain-Driven Design principles. Built in a 4-person team for the Enterprise System Integration course; I personally owned the Student Service and Notification Service end-to-end.
| Attribute | Detail |
|---|---|
| Stack | Java 21 · Spring Boot 3.4 · Spring Security 6 · Spring AMQP · RabbitMQ · PostgreSQL · SpringDoc OpenAPI · Vue 3 · Docker Compose |
| Scale | 8 microservices · 12 containers · DDD bounded contexts · per-service schemas |
| Performance | Async event-driven notification delivery · stateless JWT auth (no session store) |
| Security | Spring Security 6 · stateless JWT · per-service RBAC · principle of least privilege |
| Impact | Full integration documentation · sync/async tables · DDD models across all services |
| Repository | github.com/hashimminhas/Uni-Campus |
The Student Service handles enrollment, course assignments, and academic records; the Notification Service consumes events over RabbitMQ and delivers messages across channels. Tested with JUnit 5, MockMvc, and Mockito (@WebMvcTest slice tests) and documented exhaustively with SpringDoc OpenAPI for cross-team integration.
⚙️ DevOps & Full-Stack Microservice Platform
A production-grade Reddit-style platform deployed end-to-end on Kubernetes with full CI/CD automation and observability — my deepest hands-on exercise in cloud-native delivery.
| Attribute | Detail |
|---|---|
| Stack | Next.js (TypeScript) · Flask · PostgreSQL · Docker · Kubernetes (Minikube) · GitLab CI/CD · Prometheus · Grafana |
| Scale | 3-tier microservice architecture · multi-stage pipeline · containerized services |
| Performance | Real-time Grafana dashboards tracking latency, error rates, container resources |
| Security | JWT authentication · role-based access control across service boundaries |
| Impact | Reproducible releases via lint → test → build → deploy pipeline · full observability |
| Repository | github.com/hashimminhas/reddit-clone-devops |
End-to-end ownership of containerization, orchestration, pipeline design, and metrics — the same operational pattern that runs modern cloud-native systems at scale.
🌍 EarthScanner — Travel Planning Platform
A travel planning platform built in Phoenix (Elixir) as part of a 4-member Scrum team. Focused on concurrent request handling, transactional consistency, and behaviour-driven test coverage of the critical booking flows.
| Attribute | Detail |
|---|---|
| Stack | Elixir · Phoenix · PostgreSQL · WhiteBread (BDD) · ExUnit |
| Scale | 4-member Scrum team · sprint-based delivery · backlog grooming · code reviews |
| Performance | Concurrent user request handling · optimized booking-workflow queries |
| Security | Transactional data consistency · validated input boundaries on booking APIs |
| Impact | Regression detection across sprints via BDD suites on critical paths |
| Repository | Course project · University of Tartu |
A deliberate foray into functional concurrency with the BEAM — the architectural lessons (process isolation, supervision, message passing) carry directly into how I think about resilient microservices in any language.
🤖 Fake Review Detection — ML Classification Pipeline
Undergraduate capstone: a supervised ML pipeline detecting fraudulent product reviews on the Daraz marketplace. Benchmarked four classifiers and shipped a deployable model with full preprocessing.
| Attribute | Detail |
|---|---|
| Stack | Python · scikit-learn · XGBoost · NLTK · NumPy · pandas |
| Scale | 4 classifiers benchmarked · 10,000 TF-IDF features · Daraz marketplace dataset |
| Performance | 90.70% accuracy with Linear SVM · 90.60% precision · 90.82% recall · F1 90.71% |
| Security | N/A · classification pipeline · model artifact distribution control |
| Impact | Production-ready fraud-detection classifier with reproducible training pipeline |
| Repository | Capstone — University of Chakwal |
Complete NLP preprocessing — stopword removal, tokenization, vectorization — paired with rigorous evaluation. The kind of pragmatic ML work that earns its keep in real product pipelines.
🧩 FocusFilter — Browser Extension
A lightweight Chrome / Firefox extension that hides distracting site elements such as YouTube Shorts. Built because I needed it; open-sourced because others did too.
| Attribute | Detail |
|---|---|
| Stack | JavaScript · Chrome Extension APIs · Firefox WebExtensions |
| Scale | Cross-browser support · zero external dependencies |
| Performance | Lightweight DOM filtering · negligible page-load overhead |
| Security | Minimal-permission manifest · no telemetry · no data collection |
| Impact | Open-source utility · personal productivity tooling |
| Repository | github.com/hashimminhas/focusfilter |
Sep 2025 — Present · Tartu, Estonia
Designing, building, and presenting production-grade software projects across enterprise integration, web security, and cloud-native delivery as part of the MSc Software Engineering program (GPA 4.6/5.0). Selected projects deployed to live university infrastructure and presented at the UT ICS Student Project Contest.
Scope of work
- Architected microservice systems with strict bounded contexts, async messaging, and per-service data ownership
- Implemented stateless JWT authentication and per-service RBAC across multi-service deployments
- Built and deployed automated scraping, analytics, and notification pipelines processing 600+ records on scheduled cycles
- Designed CI/CD-ready container topologies with Docker Compose and nginx reverse-proxy routing
- Presented technical work to faculty, peers, and external attendees at the UT ICS Student Project Contest
Java Spring Boot Microservices RabbitMQ PostgreSQL Docker Kubernetes Spring Security DDD OpenAPI
2024 · Pakistan
Designed, trained, and evaluated a supervised classification system for fraudulent-review detection on the Daraz marketplace as my undergraduate capstone project.
Scope of work
- Built a complete NLP preprocessing pipeline (tokenization, stopword removal, normalization)
- Trained and benchmarked four classifiers; selected Linear SVM as best performer
- Engineered 10,000-dimensional TF-IDF feature space with reproducible training pipeline
- Reported precision, recall, F1, and accuracy with cross-validated metrics
Python scikit-learn XGBoost NLTK pandas Feature Engineering
| Recognition | Details |
|---|---|
| 🏆 UT ICS Student Project Contest — 2026 | Presented Üüriturg at the University of Tartu poster session |
| 🎓 Academic Excellence — MSc Software Engineering | GPA 4.6 / 5.0 at the University of Tartu |
| 🎮 Youth Participation Idea Competition | Certificate · Team Soda Orange · GAMEIN project · Tampere University · co-funded by the EU |
| 🚀 Live Production Deployment | Üüriturg serving 600+ listings on university infrastructure |
| 📈 Open Source Contributor | 470+ contributions across 27 repositories |
| 🌍 International Engineering Profile | Software engineering across Pakistan and Estonia |
focus:
learning:
- Distributed systems patterns (saga, CQRS, event sourcing)
- Advanced Kubernetes (operators, custom resources, GitOps with ArgoCD)
- Web security (XS-Leaks, modern CSP, post-quantum primitives)
- System design at scale (rate limiting, sharding, consistent hashing)
building:
- Üüriturg — scaling toward GDPR-compliant platform partnerships
- Backend microservice patterns in Java 21 and Spring Boot 3.x
- Personal portfolio infrastructure & developer tooling
exploring:
- Service mesh adoption (Istio, Linkerd)
- eBPF for observability and runtime security
- LLM-augmented developer productivity workflows
open_to:
- Software Engineering internships (Summer / Autumn 2026)
- Backend, DevOps, Cloud, and Security Engineering roles
- Tartu · Tallinn · Remote across the EU
- Collaborations on open-source DevOps and developer-tooling projects
contact:
email: hashim.ali@ut.ee
linkedin: linkedin.com/in/hashim-ali-400094310
github: github.com/hashimminhas
location: Tartu, Estonia
"Software engineering is the discipline of turning ambiguity into systems that hold up — long after the demo is over."