Solar cameras at slab edges that flag unclipped harnesses for builders
TL;DR Tieoffly puts battery-and-solar cameras where people fall - slab edges, scaffolds and lift shafts. It flags a worker at height without a clipped harness, or a missing edge barrier, and pings the site supervisor within a minute. Mid-size residential builders pay a monthly fee per camera.
India has no reliable official count of construction deaths. Researchers who built one from newspaper reports found 126 fatal falls between 2019 and March 2025. In 49% of them there was no fall protection, and 76% happened on residential projects. Young migrant labourers were overrepresented[1]. The authors call their sample exploratory, which tells you something by itself.
The workforce is huge and loosely supervised, with an estimated 5 crore construction workers, of whom fewer than 3.5 crore hold a welfare board card[2]. Across industries, studies find PPE is worn only about half the time, and supervision is linked to fewer injuries[3]. Can one safety officer watch every edge on a 20-storey tower? No chance.
What it does, in regulatory terms: a supervision aid for the employer's existing safety duties. It does not certify a site as safe or replace a safety officer.
Tieoffly narrows the seed's general PPE-and-behaviour idea to the one hazard that kills most often in the evidence above. Cameras with their own solar panel and cellular link clamp to columns at the working floor and move up with the slab cycle. A model trained on Indian site footage checks for a person close to an edge or on scaffold, whether a harness is worn and whether its lanyard is clipped. It also checks that edge barriers and shaft covers are in place.
Alerts go by chat app to the floor supervisor with a still image. The project manager gets a daily summary of alerts by floor, time to fix and repeat spots. Faces are blurred in stored images, and workers hear about the cameras in their own language at induction.
The user is the site safety officer or floor supervisor. The payer is the builder, or the main contractor on larger jobs, who carries the legal duty and the reputational damage after a death. Large infrastructure firms already buy camera analytics from Indian industrial-AI vendors[4]. Mid-size residential builders, where most falls in the study happened[1], rarely do. There's no public price list, so pricing per camera per month is an assumption you'll need to test.
The four Labour Codes, including the Occupational Safety, Health and Working Conditions Code, came into force on 21 November 2025[5]. That puts site safety duties under one statute and brings in a national safety board. Edge AI chips and cheap cellular modules now make a self-powered camera practical on floors that have no wiring yet.
Nobody publishes a figure for spending on construction safety technology in India. The narrower signals - residential projects made up 76% of fatal falls in the study sample[1], and global vendors have built work-at-height modules[6].
For adjacent context, the ILO counts about 330,000 deaths a year from work accidents worldwide, with construction among the most hazardous sectors[7].
One Indian industrial-AI platform runs over 300 detection use cases on existing IP cameras, including missing harnesses, and lists construction among its industries[4]. A Hong Kong-linked vendor sells PPE, danger-zone and work-at-height modules, with projects in Hong Kong[6]. Both work best on sites with fixed cameras and power.
The gap is a portable, self-powered kit priced for a residential builder running three towers.
| Difficulty | Hard (52/100) |
| Starting capital | ₹1.2–2.2 Cr |
| Time to launch | 9–12 months |
Spotting a clipped lanyard in dust, glare and rain is a lot harder than spotting a helmet. The hardware has to survive concrete pours and get moved every slab cycle. And the sale is cultural, because alerts are worthless if supervisors ignore them and some builders would honestly prefer not to have a record at all.
Yes, the hardware needs approvals before you sell, and the safety law shapes the pitch.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't treat a quiet camera as a safe site. Missed detections will happen in dust and rain, so the safety officer still owns the duty.
Want to work out the AI layer for Tieoffly? Talk to Yash →
Here's what can go south -
I think this is promising for a founder with construction site experience and hardware skills. If you can't spend weeks on dusty sites then avoid it.
Prove two things first on one tower - that supervisors fix 80% of alerts within ten minutes, and that the builder renews after three months.
My estimate is ₹1.2–2.2 Cr to launch, covering ruggedised solar cameras, model training on site footage and pilots.
Yes. Wireless cameras need WPC type approval, CCTV devices may need BIS registration, and worker images fall under the DPDP Act.
Yes. Falls from height kill construction workers everywhere, and the product travels. Check local radio, product safety and workplace privacy rules first.
The free Tieoffly validation kit has the line-by-line capital breakdown, the effort scoring, the labour-code and worker-data checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for builders and safety officers, and 5 assumptions to test. Download the kit →
Thinking about building Tieoffly or something close to it? The make-or-break work is getting supervisors to act on alerts and builders to accept the record, and that's exactly the kind of knot I like helping founders untangle. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.