[09] Case study

Slashscore AI

A self-assessment platform that helps developers understand where they shine and where they could grow. ML-powered scoring across multiple integrations, built on a serverless architecture that adapts as the integrations grow.

slashscore.ai
Industry

Dev tools · HR tech · AI

Stack

React · Node.js · ML · Serverless

Year

2021 — V1

Scope

From zero, self-assessment platform

[ 02 ] THE CHALLENGE

A fair scoring system for an unfair-by-default category

Two developers can do similar work and look identical from the outside. The hard problem was making the differences measurable in a way both sides could trust.

▸ The problem

Software engineers leave digital footprints across the platforms they use. The challenge was finding a fair system to measure those footprints in a way that was both meaningful and applicable to all developers, regardless of the kind of work they do.

The scoring couldn't game easily, couldn't disadvantage developers who work on private codebases, and couldn't reward only the loud personas of open-source. It had to feel like measurement, not surveillance, and the patterns had to be defensible to both the developer being scored and the employer reading the score.

And it had to scale. Every new integration added complexity, so the architecture had to absorb that complexity without dragging the rest of the system down.

[ 03 ] WHAT WE BUILT

A self-assessment platform with ML at the core

Multi-source data, a privacy-respecting integration model, and a serverless backbone designed for the long game.

▸ Feature

Multi-Source Data Pipeline

Integrations with the platforms developers leave digital footprints on. The pipeline that turns raw activity into measurable signal.

▸ Feature

ML-Powered Scoring

Machine learning models trained to find patterns in developer activity and translate them into fair, defensible scores.

▸ Feature

Self-Assessment Flow

Developers connect their accounts and see their score across categories. Personal insight first, hiring tool second.

▸ Feature

Strengths & Gaps View

The platform surfaces where a developer shines and where they could grow. Useful for the developer, valuable for their next employer.

▸ Feature

Serverless Architecture

Built on AWS serverless so the platform handles bursty integration loads without paying for idle compute.

▸ Feature

Privacy-First Design

Developers control which sources they connect. The platform earns trust by respecting it.

[ 04 ] OUTCOMES

A fair foundation for developer profiling

Where V1 landed.

Fair scoring

Patterns found across multiple integrations generate scores defensible to both developers and recruiters.

Scalable foundation

Serverless architecture adapts to the complexity of all possible integrations without infrastructure tax.

Developer-first product

Built as a self-assessment tool that happens to help with hiring. Not the other way around.

[ 05 ] TECH STACK

An ML + serverless stack built for integrations

React on the front, Node and ML on the back, AWS serverless underneath.

React

Frontend for the developer-facing dashboard. Designed for clarity in a category prone to dark patterns.

Node.js

Backend services for integrations, score calculation orchestration, and the API powering the UI.

Machine Learning

ML models trained on activity data from multiple integrations. Pattern detection at the core of the scoring engine.

AWS Serverless

Lambda, API Gateway, and the rest of the stack. Scales with each new integration without rewrites.

PostgreSQL

Developer profiles, integration tokens, and score history. Relational data for a domain that depends on traceable connections.

OAuth Integrations

Standard OAuth flows for connecting third-party developer platforms. Built once, applied to every new integration.

[ 06 ] MORE WORK

Other case studies

More products we've designed, built, and shipped.