Formulatrix
2026
FLO i8® PD: Automation Laboratory Software
Overview
What Does It Cost to Run High-Precision Robots on Fractured Tools?
Formulatrix builds serious biotech instruments trusted by powerhouses like Roche, Genentech, and MIT. But before these machines leave the factory floor, they have to survive a brutal validation gauntlet to prove their fluidic precision. The real cost of a fractured toolchain is not just clunky UI: it burns massive operational hours on avoidable chaos.
Problem: Laboratory validation relied on disconnected tools.
Validation relied on three completely disconnected tools stitched
Split Setup: Operators pulled files from servers and manually uploaded configs to separate pltforms.
Fragile Execution: Hardware ran as a black box: a single failure aborted the entire run.
Hand-Rolled Scoring: Raw logs had to be parsed and calculated by hand in spreadsheets.
The Bet: Goal, Hypothesis, & Validation Plan
01
Goal
Eliminate repetitive manual overhead and prevent catastrophic mid-run validation aborts.
02
Hypothesis
If we automate manual config uploads and spreadsheet math into a live telemetry system, operators can safely oversee parallel instrument runs without babysitting fragile hardware.
03
Measurement Plan
Success was tracked across three operational vectors: reduced end-to-end cycle time per instrument, zero manual calculation errors via automated parsing, and minimal line downtime caused by unmonitored hardware faults.
Solution
The feature that unifies validation so the production team doesn't have to...
Eliminating repetitive operational overhead so the production team can focus on output and throughput.
Research
We researched the automated liquid handling landscape and current validation workflows
By shadowing actual lab runs, we uncovered how messy and time-consuming the daily testing process was forcing operators to jump between multiple tools and spend hours doing manual data work.
What we learned
Through interviews and workflow analysis, I uncovered that normalized friction was the real bottleneck. Validating two instruments monthly across dozens of test variations forced operators to navigate fragmented tools, subjective scoring, and blind mid-run failures. Without systemic visibility, an inefficient workaround had become standard practice, leaving multi-hour validation runs anchored in operator guesswork rather than repeatable software.
"Running dozens of tests on these machines every month was just exhausting to manage. I'd sit there dreading every mid-run failure, knowing I'd likely have to throw out hours of progress simply because nobody could tell me what actually went wrong."
Interview participant
Operator
Prototyping & Testing
Prototyping interfaces for factory validation
Instant system-synced configuration
Protocols and configurations are instantly pulled from the backend in one click, completely eliminating manual file handling.
Frictionless real-time execution
Run validation tests smoothly with live status tracking, removing operational bottlenecks and complex multi-step setups.
Automated result delivery
Pre-calculated validation files are downloaded instantly by the system, eliminating hours of manual calculations.
Key User Insights
Diagnostics over minimalism
Operators didn't want a simplified, hidden UI. They needed persistent technical logs visible at a glance to spot micro-glitches before runs failed.Recovery, not hard stop
A minor hardware timeout shouldn't abort a 2-hour run. Operators needed ways to pause, inspect, and retry individual steps instead of starting over.
Design Decision
We designed a fault-tolerant workflow for high-stakes testing.
Building on feedback from physical test runs, we shaped solutions that protect long validation cycles from sudden halts while keeping the interface clear.
Insight 1: Catch errors before the run starts
Pre-flight validation on upload
Mismatches used to surface mid-run, wasting a full test cycle. I moved the check to the upload step and surfaced it as a toast message, a low-friction signal that stops the operator before they waste time, without blocking their screen.
Insight 2: Keep testing moving during pending software updates
State syncing & two-layer abort
When a run stopped unexpectedly, operators had no way to tell what state things were left in. I added real-time state syncing so the system always reflects where the test stopped, and a two-layer confirmation before any abort, preventing an accidental tap from discarding a run still in progress.
Insight 3: Always show where an interrupted test stopped
Granular status per channel
One failed channel used to fail the whole test, hiding which part actually broke. I redesigned results to show status per channel, using the same confirmation modal from scoring, but with copy that tells operators exactly which channel failed and which passed.
Insight 4: Keep operators working through pending updates
Fallback config popup
Blocking new tests until an update landed would have stalled operators completely. I added an info icon that opens a popup explaining the situation, with an Add Config button inside it, letting operators configure manually and keep working, safely, without waiting for the update.
Reflection
What I learned
Routine hides broken systems
Operators accepted manual workarounds as normal. My job was questioning habits everyone took for granted.
Feedback builds trust.
Silence makes operators anxious. Live status gives them the confidence to let the machine run.
