Now recruiting capture teams worldwide. Get paid to record first-person video of real work.

Apply to capture

Hard task

Egocentric Titration and Reagent Weighing Data

Titrating a sample to a colour endpoint and weighing a reagent on a balance are two bench tasks the site’s task catalog lists as their own rows, both rated Hard. This page covers those two rows: what the catalog states, how the multiview rig with handheld grippers captures them, what each episode carries and what the site does not state. We collect it as custom lab data on demand, to a written skill spec, with the row status stated exactly as the catalog lists it.
Catalog rows
Titrate to endpoint and Weigh reagent, in Wet Lab & Clinical Bench
Skill and status
Both Hard; both Available on request
Capture
Multiview rig + handheld grippers
License
Perpetual, buyer-owned

Requirements

What does a titration or weighing capture need that a general lab clip does not?

01

Two catalog rows, each its own task

Titrate a sample to a colour endpoint, and weigh a reagent on a balance, as the catalog words the two rows.

02

Hands, sample and balance in view together

Hand pose, contact events and per-frame instance masks are recorded on one clock, with a visibility flag on each joint.

03

Written accept criteria before capture

What counts as a reached endpoint or a finished weighing is written into the spec up front, because the site states no outcome definition.

04

Privacy stated plainly

Consent, anonymization and the license are stated below. This page makes no safety or compliance claim and gives no instructions.

Definition

What is egocentric titration and reagent weighing data?

Egocentric titration and reagent weighing data is first-person video of a person doing the work named in two catalog rows: titrating a sample to a colour endpoint and weighing a reagent on a balance. It is recorded from the contributor’s own point of view, with synchronized sensor streams alongside, so the frames show what the hands are doing with the sample and the reagent rather than a distant view of a bench. Searchers also phrase it as egocentric titration data, first-person reagent weighing video, a wet lab bench robot dataset or custom lab data collection. The data is the same, collected as a custom collection scoped to your task.

On this site the data is human capture: a paid, consented contributor performs the task from a written skill spec. It is not robot or teleoperation data, and the page makes no claim that any robot has been trained on it or that any robot titrates or weighs. This page is narrower than the egocentric lab pipetting and sample handling page, which covers fourteen lab rows with titration and weighing among them, and than the egocentric pharmacy and lab bench page. Here the subject is the two rows, Titrate to endpoint and Weigh reagent.

Catalog rows

Which catalog rows cover titration and reagent weighing, and what do they say?

Two rows sit in the Wet Lab & Clinical Bench environment, in the Healthcare & Life Sciences sector. "Titrate to endpoint" reads: "Titrate a sample to a colour endpoint." "Weigh reagent" reads: "Weigh a reagent on a balance." Both are rated Hard, both list the capture as Multiview rig + handheld grippers, and both are Available on request. The environment’s shared example line reads "Burette, pipettes, balance, labware, phantom for any invasive step". It is one line for every row in that environment, so it is quoted here rather than assigned to a single task.

A third row sits in the Pharmacy & Laboratory environment: "Lab protocols (pipetting, titration, sample handling)", which reads "Pipette reagents, titrate to endpoint and handle samples following the written protocol." It is rated Hard, its capture is Phone (head-mounted) / multiview rig and its status is Available on request. It is a broader row that mentions titration as one part of a written protocol, and it is covered by the lab pipetting page. The first table below reproduces all three rows exactly.

Available on request means the work is scoped and recruited to spec when a buyer asks. It does not mean a stock of titration or weighing footage exists, and this page claims no volume, hours or delivered clips for either row. The rows name a sample, a colour endpoint, a reagent and a balance. They name no chemical, no concentration, no quantity and no instrument model, so if your program needs one, it is written into the skill spec rather than assumed.

Why it is hard

Why are titration and reagent weighing listed as Hard?

The site’s answer is the Skill column. The catalog rates both rows Hard, and that is all "Hard" means here. This page imports no outside rubric and states no failure rate, tolerance or claim about how skilled any worker is. What the site does say is in the row text. The titration row is defined by an endpoint that is a colour, so the end of the task is a visual state of the sample, not a count or a position. The weighing row is defined by a reagent on a balance. The site publishes no reading, no limit and no step list for either.

The site’s lab copy adds two things that bear on capture. The first is that small items sit in or near the fingers for the whole task, so the hands are the data, which is why hand pose carries a per-joint visibility flag for occluded fingers. The second is that depth degrades on wet and reflective surfaces and that a confidence map is delivered so low-confidence regions can be masked. The challenge table below maps each difficulty to the stream or label the site already states, and says Not stated where nothing is.

Capture

How are titration and reagent weighing captured?

The catalog lists one capture mode for both Wet Lab & Clinical Bench rows: Multiview rig + handheld grippers. The site describes it as a staged working set recorded by a multiview rig while the operator performs the task with handheld grippers, so demonstrations map onto a robot end effector. This page does not describe these rows as head-mounted phone capture, and it does not describe them as stereo capture. Stereo is a separate offer on this site, agreed in a spec and confirmed on a Spec Pilot, and neither row states it.

The site also describes the wider egocentric rig: ten synchronized streams (4K head camera, two wrist cameras, depth, IMU, 3D hand and body pose, instance segmentation and action labels), hardware-triggered onto one PTP clock domain and aligned under 2 ms. Which of those streams apply to a staged gripper setup is not stated, and neither is how hand pose is recorded when the operator holds a gripper. Confirm both in the skill spec. The broader Pharmacy & Laboratory row uses the head-mounted phone and multiview rig described on the environment page. Force and tactile readings are not among the listed streams.

Episodes

What does each titration or weighing episode contain?

Each attempt is delivered as one episode. The standard episode carries action segments, 3D hand pose for both hands with a per-joint visibility flag, contact events that mark when a hand meets an item, an objects list, and per-frame instance masks stored at the depth resolution of 848×480. Hand pose is 21 joints per hand, 3D, in metres in the head-camera frame. For these rows an episode might cover one weighing, one titration or a longer run of both; where each episode starts and ends is written into the spec, because the site sets no episode length.

Action segments divide an episode into start- and end-timestamped spans, each carrying start_ns, end_ns, verb and noun. The site lists no label for the outcome of a step, such as whether an endpoint was reached or what a balance displayed, no label for a colour change and no reading, so ask for any such label in the spec.

Validation

How is the data validated and accepted?

Validation runs against signed accept criteria written into the spec before capture, and a reject log ships with each batch. Episodes are machine-checked against stated reject criteria, such as a sync breach, the manipulated object out of frame for more than 15 percent of the episode, or more than 35 percent of depth pixels in the manipulation region invalid, and a human reviewer then checks each one. Only footage that clears the criteria is billed as a validated hour.

For these rows, the spec has to say which item counts as the manipulated object at each step and where the manipulation region is, because the sample, the reagent and the balance are different items at different moments. What counts as a reached endpoint or a finished weighing is also written down, since the site states no outcome definition. Rejected items are recaptured at our cost, and the site states no accuracy figure for step labels, so what counts as a pass for a step boundary is set in the accept criteria.

Delivery and license

Which formats and license apply?

Delivery is in the format your stack already reads: RLDS, LeRobot, WebDataset, HDF5, zarr or Rerun .rrd, or raw streams with metadata, with readable converter source alongside. The license is perpetual, irrevocable and worldwide, buyer-owned and non-exclusive by default, with exclusivity available as a separate term. The executed agreement controls.

Official documentation for the RLDS, LeRobot and WebDataset formats is linked in the sources below. MCAP delivery is a separate offer described on the egocentric stereo MCAP page, and nothing on this page promises an MCAP file for either row.

Neighbouring pages

How does this page differ from the other lab pages?

Several pages touch the same bench, and none has titration or reagent weighing as its main topic. The egocentric lab pipetting and sample handling page lists the two rows among fourteen and centres on pipetting and sample tubes. The stereo version covers the same lab family on the stereo and MCAP side. The egocentric pharmacy and lab bench page spans medication dispensing and labeling as well as lab work, and the egocentric microbiology slide and plate streaking page is about two other Hard rows, Prepare a slide and Swab + streak plate.

The catalog also has a Medium row, "Pour + measure reagent" ("Pour to a graduated cylinder mark."), which is a neighbouring row and not part of this page. The comparison table below sets each page against this one.

Capture and privacy

How are contributors and their work protected?

Every contributor signs a written release, and each delivered episode links to it through consent artefact IDs. At ingest, automated detection finds faces, plates and screens and blurs them irreversibly, and speech is scrubbed of names, addresses and card numbers. A human reviewer confirms that no un-consented person is identifiable in any frame, and the anonymization record ships with the data. Firsthand does not capture at a customer’s own facility, and which setting a program uses is agreed in the skill spec before capture begins.

Container labels and the readout of a balance in frame are not listed as blur categories, so raise them in the spec if they matter. The catalog notes that the Pharmacy & Laboratory work requires regulatory compliance, and this page makes no regulatory, laboratory-practice, biosafety, clinical or diagnostic claim of its own. It gives no instructions for any step. None of this is legal advice.

Speed and scale

How fast and at what scale can a custom titration and weighing dataset be collected?

From signed spec to first validated batch, the published timeline is a median of 48 hours where coverage is deep and three to four weeks where it is thin and contributors need to be recruited first. Both rows are Available on request, so the range that applies is confirmed when the spec is quoted rather than assumed. After a first batch, on-demand top-ups run against the same accept criteria, so you can add steps without rewriting the spec.

Titration and reagent weighing are custom collection at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs. The site states no provider count, hours or throughput for these rows, and none of its off-the-shelf dataset descriptions names them, so treat them as a custom collection.

Getting started

How do you start a pilot?

Describe the steps and where each starts and ends, what you expect to see as the sample, the reagent and the balance, what counts as a finished step and the failure or variation cases your pipeline is missing. That becomes a written skill spec, quoted for reach, timeline and price before any capture begins. Our skill spec guide walks through how to write one.

Most programs begin with a Spec Pilot, a small, fast first batch that proves the spec and the accept criteria on real footage before volume scales. Collections are billed per validated hour. Counts of sites, episodes and hours are set in the skill spec.

Capture setup

What is the capture setup for titration and reagent weighing?

The capture setup as stated on this site. Counts of sites, contributors, episodes and hours are set in the skill spec.
PartAs stated on this site
Capture mode (both Wet Lab rows)Multiview rig + handheld grippers, as the catalog lists it
What that meansA staged working set recorded by a multiview rig while the operator performs the task with handheld grippers, so demonstrations map onto a robot end effector
Capture mode (Pharmacy & Laboratory row)Phone (head-mounted) / multiview rig, as the catalog lists it
Streams (egocentric rig)Ten: 4K head camera, two wrist cameras, depth, IMU, 3D hand and body pose, instance segmentation, action labels
SyncHardware-triggered onto one PTP clock domain; aligned under 2 ms
Hand pose21 joints per hand, 3D, in metres in the head-camera frame; per-joint visibility flag
StereoNot stated for these rows; a separate offer, agreed in the skill spec and confirmed on a Spec Pilot
Which streams apply to the gripper setupNot stated; confirmed in the skill spec

Every collection runs through the same seven-stage end-to-end custom collection pipeline.

Live catalog rows

Stated row × description × skill × capture × status

The two Wet Lab rows and the broader Pharmacy & Laboratory row as the task catalog states them, reproduced without change.
TaskDescriptionEnvironmentSkillCaptureStatus
Titrate to endpointTitrate a sample to a colour endpoint.Wet Lab & Clinical BenchHardMultiview rig + handheld grippersAvailable on request
Weigh reagentWeigh a reagent on a balance.Wet Lab & Clinical BenchHardMultiview rig + handheld grippersAvailable on request
Lab protocols (pipetting, titration, sample handling)Pipette reagents, titrate to endpoint and handle samples following the written protocol.Pharmacy & LaboratoryHardPhone (head-mounted) / multiview rigAvailable on request

Neighbouring pages

Page × main topic × where titration and weighing appear

How the nearest lab pages differ from this one. Only what each page states is listed.
PageMain topicWhere titration and weighing appear
This pageThe Titrate to endpoint and Weigh reagent rows, egocentricThe whole page
Egocentric lab pipetting and sample handling dataPipetting, liquid handling and sample tubes across fourteen lab rowsTwo of fourteen rows
Stereo lab pipetting and sample handling dataThe same lab family on the stereo and MCAP sideAmong the lab rows
Egocentric pharmacy and lab bench dataPharmacy dispensing, labeling and lab bench workAmong the lab rows
Egocentric microbiology slide and plate streaking dataPrepare a slide and Swab + streak plateNot covered
Hard-task data collectionThe hub of rows the catalog rates HardBoth rows are in the Hard-row table and link here

Capture challenges

Bench challenge × what goes wrong × stated stream or label

Each capture challenge mapped to a stream or label this site already states. Where nothing is stated, the row says so.
Bench challengeWhat goes wrongStated stream or label
Hands cover the item being heldFingers hide the sample or the reagent3D hand pose with per-joint visibility flags; per-frame instance masks; reject if the object is out of frame for more than 15% of an episode
Hand meets sample or balanceThe hand and the item touch in quick successionContact events mark when a hand meets an item; no finer contact label is stated
Wet and reflective surfacesDepth readings degrade on a wet or shiny surfaceThe site states that depth degrades on wet and reflective surfaces and that a confidence map is delivered to mask low-confidence regions; reject if more than 35% of depth pixels in the manipulation region are invalid
Reaching a colour endpointThe end of the titration is a colour, not a countAction segments and instance masks; not stated: no colour label, no endpoint label and no colour-accuracy figure
Reading the balanceA display or a reading may be small or out of viewNot stated: no label for a displayed value and no reading is listed
Step boundariesIt is unclear where one step ends and the next beginsAction segments, one per step as written in the spec; no accuracy figure for step labels is stated
Grip and handling forceHow hard an item is heldNot stated: no force or tactile stream is listed

Delivered vs not stated

What is delivered and what is not stated

Delivered items are stated elsewhere on this site. Items marked not stated are not promised on this page.
ItemStatusDetail
Action segmentsDeliveredstart_ns, end_ns, verb, noun, one per step
3D hand poseDelivered21 joints per hand; per-joint visibility flag for occluded fingers
Contact eventsDeliveredThe moment a hand meets an item
Per-frame instance masksDeliveredStored at 848×480
Consent and anonymization recordsDeliveredConsent artefact IDs per episode; anonymization record shipped with the data
FormatsDeliveredRLDS, LeRobot, WebDataset, HDF5, zarr, Rerun .rrd or raw streams with metadata
Samples, reagents and balance typesNot statedAgreed in the skill spec
Endpoint, colour or reading labelsNot statedNo such label is listed
Cycle times and throughputNot statedNo such figure is listed
Number of sites, episodes and hoursNot statedSet per program in the skill spec
Force or tactile readingsNot statedNot among the listed streams
Regulatory, safety or clinical claimsNot statedThis page makes none

Example footage

What does first-person capture look like?

One real clip already published on this site, shown as a format and rig example: ten seconds of first-person home cooking with faces blurred. It does not show titration, weighing or any lab work.

egocentric_cooking_10s.mp4 · first-person view · faces blurred · 1080 × 1080 web preview · format example, not a titration or weighing clip

No titration or weighing clip is published on the site, so this is the nearest first-person clip, labelled as a format example. Request a sample to see delivered files. See sample data outputs, request an egocentric data sample.

FAQ

Egocentric titration and reagent weighing data, answered.

What is egocentric titration and reagent weighing data?
Custom first-person video of a paid, consented contributor titrating a sample to a colour endpoint and weighing a reagent on a balance, captured to a written skill spec.
Which catalog rows cover it, and what is their status?
Two rows in Wet Lab & Clinical Bench: "Titrate to endpoint" ("Titrate a sample to a colour endpoint.") and "Weigh reagent" ("Weigh a reagent on a balance."). Both are rated Hard and both are Available on request. This page claims no volume, hours or delivered clips.
Is it captured with a head-mounted phone or in stereo?
The catalog lists Multiview rig + handheld grippers for both Wet Lab rows. Head-mounted phone capture is listed for the broader Pharmacy & Laboratory protocol row. Stereo is not stated for either; it is a separate offer agreed in the skill spec and confirmed on a Spec Pilot.
Does the data include endpoint, colour or balance-reading labels?
Not stated. The site lists action segments, hand pose, contact events and instance masks, and no label for a colour change, a reached endpoint or a displayed reading. Ask for any such label in the spec.
Is there an off-the-shelf titration or weighing dataset?
Not stated. These rows are a custom collection, at scale through a network of thousands of capture providers, for a consistent, recurring data flow to labs.
Will faces or screens be visible?
Faces, plates and screens are blurred irreversibly at ingest, and a human reviewer confirms that no un-consented person is identifiable. Whether container labels or a balance display in frame are protected beyond the license is not stated.
Who owns the data and in which formats does it arrive?
The buyer owns it under a perpetual, non-exclusive commercial license by default. Formats include RLDS, LeRobot, WebDataset, HDF5, zarr and Rerun .rrd.
How fast can a fast, on-demand titration and weighing dataset arrive?
A median of a median of 48 hours from signed spec to first validated batch where coverage is deep, and three to four weeks where it is thin. Both rows are Available on request, so the range is confirmed when the spec is quoted.

Sources

Where are the delivery formats documented?

Request a custom titration and reagent weighing dataset.

Tell us the steps and where each starts and ends, what you expect to see as the sample, the reagent and the balance, whether an episode covers one task or a run, and the angles you want. We will quote reach, timeline and price against a written spec.