A lightweight applicant tracking system built for startup founders who are doing their own recruiting. Scheduling, messaging, scorecards, and pipeline management, all in one place, built to feel simple even when the workflows underneath are not.
Context
Most applicant tracking systems are built for recruiting teams. They assume you have a dedicated recruiter, an HR coordinator, and someone managing interview logistics full time. At early-stage startups, that person doesn't exist. The founder is the recruiter, the hiring manager, and the interviewer, all at once.
These founders were cobbling together hiring workflows from spreadsheets, email threads, shared Google Docs, and calendar invites. Candidates would fall through the cracks. Interview feedback lived in Slack messages that disappeared after a week. Nobody had a clear picture of where things stood.
The existing ATS market didn't serve them. Enterprise tools like Greenhouse and Lever were too expensive and too complex. Lighter tools still required significant setup and assumed recruiting vocabulary that founders didn't have. AngelList had a unique opportunity: we already had the candidates. We just needed to give founders the tools to actually manage the process.
Discovery & Research
Before designing anything, we needed to map the real workflow. Not the idealized version, but the messy reality of how a founder with no recruiting background moves a candidate from "interesting" to "hired."
Journey mapping the founder hiring workflow, from first applicant to signed offer
Strategy & Framing
The research revealed four core jobs that every founder was trying to accomplish, usually with a patchwork of tools that didn't talk to each other. These became the pillars of Track.
Founders need to know where every candidate stands without clicking into individual profiles or cross-referencing a spreadsheet. A single, visual pipeline that shows who's in review, who's interviewing, and who's waiting on a decision.
Scheduling was the single biggest time sink. Track needed integrated calendar support that let founders share availability and let candidates self-book, eliminating the five-email chain it usually took to lock in a time.
Interview scorecards that a founder can fill out in 90 seconds, not 10 minutes. Lightweight enough to actually get used, but structured enough to support real comparisons when it's time to make a decision.
Messaging that lives inside the hiring flow, not scattered across personal inboxes. Templates for common touchpoints, automated status updates, and a clear thread for every candidate so nothing gets lost.
Design Process
The core design challenge was clear from the start: an ATS is inherently complex. It touches scheduling, communication, evaluation, and decision-making. The goal was not to remove that complexity but to organize it so well that founders never felt overwhelmed.
The first challenge was structural. Traditional ATS tools organize around features: a scheduling section, a messaging section, an evaluation section. But founders don't think in features. They think in candidates. So we organized everything around the candidate card. Every action you could take, schedule, message, score, or move to the next stage, was accessible from the candidate's context.
We designed every surface with layers. The pipeline board shows the minimum you need: name, role, stage, and days in stage. Click into a card and you get the full profile, interview history, and scorecard data. Click deeper and you reach scheduling details, message threads, and attached files. Nothing is hidden, but nothing is forced on you either.
This approach was tested extensively. Early prototypes showed everything at once, and founders told us it felt like a cockpit. The layered version tested dramatically better for both speed and confidence in decision-making.
We ran usability tests at every major milestone. The kanban board went through four iterations based on founder feedback. The scorecard flow was completely rebuilt after the first round of testing revealed that founders wanted to score immediately after an interview, not later as a separate task.
The scheduling integration was the most technically complex piece. We tested three different models before landing on a self-serve booking link that founders could send directly from a candidate's card, with their calendar availability synced automatically.
Key Design Decisions
We chose a kanban board as the default view because it maps directly to how founders already think about candidates: a sequence of stages from applied to hired. Every founder we interviewed was already drawing this on a whiteboard or building it in a spreadsheet. We just made it the real thing, with drag-and-drop to advance candidates, inline actions, and smart defaults for stage names that match how early-stage hiring actually works.
Most lightweight ATS tools punt on scheduling and tell you to use Calendly. We built it in. When a founder moves a candidate to the interview stage, they can send a booking link in one click. The candidate picks a time, it lands on the founder's calendar, and Track updates the pipeline automatically. No copy-pasting links, no switching tabs, no missed confirmations. This single feature cut scheduling time by over 70% in our testing.
Enterprise scorecards have 15 dimensions and take 10 minutes to fill out. Founders won't do that. So we designed a scorecard with three to five criteria, a simple thumbs-up, neutral, or thumbs-down rating, and one free-text field for notes. The entire thing takes under 90 seconds. Completion rates went from 23% with our first detailed version to 84% with the simplified design.
Outcomes
Track launched as a free ATS for AngelList users, giving startup founders a complete hiring workflow without the price tag or complexity of traditional tools. It shipped incrementally over 18 months, with the pipeline, scheduling, messaging, and scorecards rolling out as connected modules.
Reflection
The biggest lesson from Track was about audience. We were designing an ATS, but our users weren't recruiters. They were founders, engineers, operators, people who had never used an ATS before and didn't want to learn one. Every piece of recruiting jargon we removed, every default we set correctly, and every workflow we shortened made the difference between a tool that got adopted and one that got abandoned.
I also learned the power of saying "not yet" instead of "no." There were dozens of features that made sense on paper, things like advanced reporting, multi-team permissions, and custom pipeline automations. But adding them too early would have destroyed the simplicity that made Track work. We kept a disciplined roadmap and let usage data tell us when founders were ready for more.
Working on Track reinforced something I keep coming back to: the best design work often looks obvious in retrospect. A kanban board, a booking link, a simple scorecard. None of these are revolutionary ideas. But getting the details right, the defaults, the progressive disclosure, the micro-interactions that make drag-and-drop feel satisfying, that's where the craft lives.