Impact drivensoftware engineer

Leon IngmanLund, Sweden · currently Barcelona

Rendezvous

Maritime intelligence for the Baltic shadow fleet. Detects and scores the vessels that move sanctioned oil with their transponders switched off.

2025

Rendezvous answers a question that is hard by construction: which ships are hiding? The platform sweeps the Baltic operational theatre for AIS positions on a continuous cycle, enriches each vessel with encounter data, and scores it with an Elkan-Noto positive-unlabelled XGBoost classifier — a framing chosen because there is no clean set of confirmed innocent vessels to learn against, only a handful of known offenders and a sea of unlabelled traffic. Every score carries a SHAP explanation, so an analyst sees the evidence rather than a number. The backend is FastAPI over PostgreSQL and Redis, deliberately real-time with no historical retention beyond per-table TTLs; the operator console is Next.js with a live Mapbox tracker, an intelligence feed, and shadow-fleet alerts.

Built with

  • FastAPI
  • XGBoost
  • Mapbox
  • Next.js
Open project

Solar Neighbours

Finds the neighbours a shared solar plant should call first. Ranks every building around it by how much of its power use falls in sun hours.

2026

Shared solar in Spain lets any building within reach of a plant take a share of its power, yet plant owners still find participants door to door, and whatever the neighbours don't use is sold to the grid for almost nothing. Solar Neighbours takes an address or a plant and ranks every building around it from open data: Catastro footprints and building use, hourly PV output from PVGIS, consumption shapes per tariff from REE ESIOS, OMIE wholesale prices and the AMB rooftop-solar map. Each neighbour gets its share of the plant, its saving at the participant price, the CO₂ avoided and one sentence on why to call it; buildings that already carry panels in OpenStreetMap are left out. An owner selects a community, and a Claude agent emails them a follow-up plan. Built in 24 hours at the Norrsken Fixathon 'Fixing the Paris Agreement' in Barcelona — FastAPI and PostGIS on Railway, a Next.js and Mapbox front end on Vercel, and a background loader that pulls each source in on demand.

Built with

  • FastAPI
  • PostGIS
  • Next.js
  • Mapbox
Open project

Monti

An AI procurement platform for Swedish consulting firms. Turns a public tender into a scored opportunity and a drafted bid.

2026

Consulting firms lose bids to the arithmetic of effort: writing a proposal costs days, so the wrong tenders get answered and the right ones get missed. Monti uploads a tender, extracts the skills it demands, matches them against the firm's consultants, and returns a fit and win verdict so the pipeline can be sorted by expected return. Once a bid is committed, the system generates the document against the firm's own Word template, filling what it can and reserving the rest as assignable tasks the team resolves in a shared workspace. Built as four services — a Kotlin and Spring Boot API owning auth and per-organisation isolation, a Python worker running all LLM jobs off a Postgres-backed queue, and Next.js marketing and console surfaces.

Built with

  • Kotlin
  • Spring Boot
  • Python
  • Next.js
Open project

Driver

A marketplace for repo fixes. Companies connect their GitHub repositories, developers browse open issues and submit changes — previewed and approved before a pull request is ever opened.

2026

Driver turns open issues into a marketplace. A company links a repository through a GitHub App, developers browse the available issues and open a fix window — Claude-assisted or manual — and implement a solution against a Monaco editor. On submit, Driver spins up an e2b sandbox to build a live preview of the change, so the company reviews working software rather than a diff. Only once the preview is approved does Driver open the pull request and reveal the source. Built as a monorepo: a Next.js frontend, a Hono API that owns all GitHub communication and sandbox orchestration, and a shared PostgreSQL layer through Prisma.

Built with

  • Next.js
  • Hono
  • Claude
  • e2b
Open project

MyTrialsFinder

A platform matching cancer patients with active clinical trials across Europe. Developed by Lunicore Student Consulting for Lunds University hospital and the County of Skåne.

2024

Built as part of a Lunicore Student Consulting engagement for Lunds University Hospital and Region Skåne. The platform aggregates active clinical trials across Europe and uses intelligent matching to connect patients with relevant studies, reducing the time between diagnosis and access to cutting-edge treatments. Developed with a Spring Boot backend and a React Native mobile frontend, with a strong focus on accessibility and medical data compliance.

Built with

  • Spring Boot
  • React Native
  • Healthcare
Open project

Risk

A crowdfunding platform opening pre-seed financing to the public. Democratising early-stage investment through a cross-platform experience.

2024

Risk is an end-to-end platform that democratises pre-seed investment by letting everyday people back early-stage startups through structured crowdfunding rounds. The product includes a Next.js web application and a SwiftUI iOS companion app, both powered by a shared Firebase backend with real-time investment tracking, KYC flows, and campaign analytics. Designed from first principles to feel as intuitive as consumer banking.

Built with

  • Next.js
  • SwiftUI
  • Firebase
Open project

Todd

A pet-management app built with Swift and Firebase. Helps owners track health, appointments, and daily care routines for their pets.

2023

A personal project exploring the SwiftUI ecosystem end-to-end. Todd helps pet owners maintain a comprehensive health log covering vet appointments, medication schedules, feeding routines, and weight tracking. Data is persisted on Firebase with an offline-first approach for reliability, and the UI is built around a calm, friendly visual language that reflects the app's purpose.

Built with

  • Swift
  • SwiftUI
  • Firebase

Amfora

A wine catalogue website for an e-commerce wine store. Built with a focus on clean product presentation and seamless purchasing experience.

2024

A frontend-first project for a boutique wine retailer, designed to elevate the online shopping experience for a curated catalogue. Heavy emphasis on UI design — smooth page transitions, refined typography, and a visual identity that communicates quality without excess. The architecture prioritises fast load times and a frictionless path from discovery to checkout.

Built with

  • E-commerce
  • Web
  • UI Design
Open project

Where I've been

Aug 2026 — present

Operations Intern

Operations at Norrsken House Barcelona, the impact hub housing founders building companies that solve real problems. Day to day I support front of house — welcoming members and visitors and handling their requests — run operational checks on rooms, equipment and shared spaces, and manage back-of-house logistics covering packages, lockers and supplies. I log issues for the Operations Lead, and treat recurring friction as something to fix rather than absorb: spotting the patterns behind repeat problems and proposing process improvements instead of maintaining the status quo.

Nov 2025 — present

Founding Engineer

Baltic Maritime Intelligence

Building Rendezvous, a maritime intelligence platform that detects and scores the Baltic shadow fleet — vessels moving sanctioned oil that switch off their transponders at the right meridian. I built the detection pipeline end to end: a FastAPI service that ingests AIS data on a continuous cycle, enriches it with vessel encounter data, and scores every vessel in the theatre with a positive-unlabelled XGBoost classifier, with SHAP explanations so an analyst can see why a ship was flagged. On top of it sits a Next.js operator console with a live Mapbox tracker, an intelligence feed, and alerting, backed by PostgreSQL and Redis.

Aug 2026 — present

Founding Engineer

AI Procurement Platform

Building Monti, an AI procurement platform for Swedish consulting firms. A firm uploads a public tender and Monti extracts the required skills, suggests a team from its consultant pool, and returns a fit and win verdict — then auto-generates the bid document against the firm's own Word template, filling what it can and reserving the rest as tasks the consultants resolve together in a shared workspace. I architected and built the system as four services: a Kotlin and Spring Boot API owning auth and multi-tenant isolation, a Python worker that runs every LLM job off a Postgres-backed queue, and two Next.js surfaces for the marketing site and the authenticated console.

May 2025 — Aug 2026

Software Development Consultant

Working as a student consultant on software development engagements, delivering full-stack solutions for clients across healthcare, finance, and enterprise sectors. Lunicore connects top engineering students with companies seeking high-quality, cost-effective development capacity.

August 2024

Project Consultant

Contributed to a project initiative at Doconomy, a fintech company building tools that help banks and consumers measure and reduce the carbon footprint of financial transactions. Worked closely with the product team during a focused one-month engagement.

Aug 2022 — Aug 2023

Junior Business Support

Provided business and administrative support at one of Sweden's leading corporate law firms, specialising in M&A, capital markets, and private equity. Worked across legal teams on case preparation, document management, and client correspondence in a fast-paced professional environment.

Jul 2020 — Aug 2020

Summer Internship YouthLab

Summer placement at :Part AB through the YouthLab initiative, a programme designed to give young people hands-on exposure to startup culture and digital product development. Gained early experience in agile workflows, team collaboration, and rapid prototyping.

Who I am

Machine learning at LTH. Nonprofit founder. Minimalist, timeless systems.

Education

Master of Science in Engineering
Computer Science and Engineering
Specialisation: Machine Learning
LTH Faculty of Engineering · Lund · Aug 2023 — Jun 2028 (expected)
Natural Science
Kungsholmens Gymnasium · Stockholm · Aug 2019 — Jun 2022
Computer Science
Pre-College Program
Brown University · Providence, RI · Summer 2019
Topics in Physics
Summer Pre-Collegiate Program
Stanford University · Stanford, CA · Summer 2018

Voluntary & extracurricular

Served as Vice Head of Economics in Spexet, one of the performing companies of Lundakarnevalen — Scandinavia's largest student carnival, held every four years in Lund. Responsible for budgeting, financial planning, cost tracking, and economic reporting across the production, through to the financial close after the 2026 carnival.

Visit Spexet, Karnevalen

Built and maintained the digital infrastructure for Lundakarnevalen as part of Satell-IT, the carnival's technology division. Worked with a team of student developers to deliver web platforms, ticketing systems, and internal tools supporting the organisation's operations in the run-up to the 2026 carnival.

Visit Satell-IT, Karnevalen

Participating as a sewist in Lundaspexarna, the renowned Lund student theatre known for its elaborate annual productions and intricate hand-crafted costumes. Contributing to the costume design and production process alongside a dedicated creative team.

Visit Lundaspexarna

Coordinated social events and activities at Malmö Nation, one of Lund's thirteen student nations. Managed logistics, scheduling, and on-the-ground coordination for a wide range of events throughout 2025, from weekly gatherings to larger student festivals.

Visit Malmö Nation

Served as Vice President of Datasektionen (the D-Guild), the student union representing computer science and information technology students at LTH. Led the executive board in initiatives spanning career development, student welfare, social programming, and inter-guild collaboration.

Visit The D-Guild at TLTH

Elected to the national board of the Swedish Federation of Young Scientists, the umbrella organisation for youth science associations across Sweden. Contributed to national science communication efforts, policy positions, and the coordination of member associations.

Visit Sveriges Unga Forskares Förbundsstyrelse

Founded KYSS to inspire students at Kungsholmens Gymnasium to explore scientific research beyond the classroom. Organised guest lectures, university lab visits, and science competitions, and grew the association to become a recognised member of the Swedish Federation of Young Scientists.

Visit Kungsholmens Young Scientists Society

Led the student union representing approximately 1,400 students at Kungsholmens Gymnasium as President of the Executive Board. Managed a team of 20+ elected representatives and drove initiatives across student welfare, educational policy, and campus culture.

Visit Kungsholmens Gymnasiums Student Union

What I build with

Languages

  • Java
  • Kotlin
  • C#
  • JavaScript
  • TypeScript
  • Swift
  • Python
  • Haskell
  • R
  • Clojure
  • Dart
  • HTML / CSS
  • SQL
  • SystemVerilog

Frameworks & tools

  • PyTorch
  • scikit-learn
  • XGBoost
  • NumPy
  • pandas
  • Next.js
  • React
  • React Native
  • React Router
  • Flutter
  • SwiftUI
  • Firebase
  • Neon
  • Spring Boot
  • FastAPI
  • Docker
  • Redis
  • MongoDB
  • PostgreSQL
  • SQLite
  • AWS
  • Vercel
  • Clerk
  • Jenkins
  • Figma
  • Photoshop
  • R-Studio

Let's buildsomething together.

Leon.ingman@gmail.com

Leon Håkansson Ingman

Fullstack & Machine Learning Engineer

Barcelona, Spain | +46 70 875 01 80 | leon.ingman@gmail.com | leoningman.se | github.com/leoningman | linkedin.com/in/leon-håkansson-ingman-b37325182

Summary

Fullstack engineer and MSc student in Computer Science and Engineering at LTH, specialising in machine learning. Founding engineer at two startups, taking AI systems from prototype to production while working across sales and marketing. Experienced in organisation management from leading student unions, boards and carnival finances.

Education

Lund University, Faculty of Engineering (LTH), Lund, Sweden

Aug 2023 – Jun 2028 (expected)

Master of Science in Engineering, Computer Science and Engineering. Specialisation: Machine Learning

Pre-college programs: Brown University (Computer Science), Stanford University (Physics)

2018 – 2019

Experience

Founding Engineer, Monti

Aug 2026 – Present
  • Architected a four-service AI procurement platform: Kotlin and Spring Boot API with multi-tenant isolation, a Python LLM worker on a Postgres-backed queue, and two Next.js frontends.
  • Built LLM pipelines that extract required skills from public tenders, match them to a firm's consultants and return a fit and win verdict to prioritise bids.
  • Automated bid drafting against each firm's own Word template, turning unfilled sections into assignable tasks.

Operations Intern, Norrsken House Barcelona, Barcelona, Spain

Aug 2026 – Present
  • Work on operations and automations for the startup impact hub, turning recurring issues into better processes.

Founding Engineer, Rendezvous AB

Nov 2025 – Present
  • Built an end-to-end FastAPI pipeline that ingests AIS positions on a continuous cycle, enriches them with encounter data and risk-scores over 10,000 vessels every day.
  • Framed shadow-fleet detection as positive-unlabelled learning (Elkan–Noto with XGBoost), with SHAP explanations showing analysts why each vessel was flagged.
  • Shipped a Next.js operator console for 256 users with a live Mapbox tracker, on PostgreSQL and Redis.

Software Development Consultant, Lunicore Student Consulting AB, Lund, Sweden

May 2025 – Aug 2026
  • Developed MyTrialsFinder for Lund University Hospital and Region Skåne, a Spring Boot and React Native app on the App Store matching cancer patients with clinical trials across Europe.

Project Consultant, Doconomy AB, Stockholm, Sweden

Aug 2024
  • Supported the product team of a climate fintech measuring the carbon footprint of financial transactions.

Junior Business Support, Advokatfirman Cederquist KB, Stockholm, Sweden

Aug 2022 – Aug 2023
  • Supported M&A, capital markets and private equity teams with case preparation and document management.

Projects

Solar Neighbours | FastAPI, PostGIS, Next.js, Mapbox, Claude | solarneighbours.today

2026
  • Ranks buildings for a shared solar plant from open data; a Claude agent emails owners. 24h Norrsken Fixathon.

Driver | Next.js, Hono, PostgreSQL, Prisma, Claude, e2b | github.com/LeonIngman/driver-dev

2026
  • Marketplace for fixing companies' GitHub issues, with live previews in e2b sandboxes before any pull request.

Risk | Next.js, SwiftUI, Firebase | riskfin.tech

2024
  • Pre-seed crowdfunding platform with web and iOS clients, real-time investment tracking and KYC flows.

Leadership and Volunteering

Vice Head of Economics, Spexet, Lundakarnevalen. Budgeting and financial reporting

Dec 2025 – May 2026

Software and Web Developer, Satell-IT, Lundakarnevalen

Dec 2025 – Apr 2026

Vice President, The D-Guild at TLTH. Computer science student union

Jan 2024 – Jan 2025

President, Kungsholmens Gymnasium Student Union. About 1,400 students

Mar 2021 – Mar 2022

Founder and President, Kungsholmens Young Scientists Society

May 2021 – May 2022

Skills

Programming: Python, TypeScript, JavaScript, Kotlin, Java, Swift, SQL, C#, Haskell, R

Machine Learning: PyTorch, scikit-learn, XGBoost, SHAP, pandas, NumPy, LLM pipelines

Frameworks: Next.js, React, React Native, SwiftUI, Spring Boot, FastAPI, Hono

Data and Infrastructure: PostgreSQL, Redis, MongoDB, Firebase, Prisma, Docker, AWS, Vercel

Languages: Swedish, English, Spanish

Project write-ups and live work at leoningman.se