Mecka AI acquires Docula to process motion data
Mecka AI, a New York and Toronto-based startup building data infrastructure for humanoid and industrial robots, acquired three-person Vancouver startup Docula earlier this year, according to BetaKit, with deal terms undisclosed. Docula's team, whose prior product processed and normalized medical-billing records, joined Mecka to help handle what cofounder Mark Grinev called "massive volumes of data" the company collects by sending iPhones and custom cameras to contributors across 12 countries. The acquisition follows a previously unannounced $60 million Series A that Fortune reports was led by Framework Ventures, with Menlo Ventures, SV Angel, and ex-Google DeepMind researcher Ted Xiao participating; Mecka is using the funding to scale toward a projected $100 million annual run rate. The deal illustrates how well-funded physical-AI data providers are absorbing regulated-data expertise as motion-data pipelines scale.
This is a three-person tuck-in, but it reads differently next to Mecka's finances: BetaKit's acquisition story does not mention that Mecka had already disclosed a $60 million raise (reported separately by Fortune) and is chasing a $100 million annual run rate, which reframes the deal as a well-capitalized data company buying niche compliance-grade processing talent rather than a distressed pickup. For practitioners, the durable pattern is that physical-AI training-data providers are recruiting people who built auditable, regulated-data pipelines, healthcare billing in this case, because schema normalization, error correction, and provenance tracking map directly onto motion-capture data quality problems.
What happened
Toronto- and New York-based Mecka AI acquired Vancouver's Docula earlier this year, BetaKit reports; all three Docula team members joined Mecka, and deal terms were not disclosed. Docula had built an AI data-processing engine for medical billing that ingested records, normalized codes, ran edits, and produced defensible reports, a capability Mecka special-projects lead Mark Grinev told BetaKit was useful because Docula was "dealing with massive volumes of data" even though its prior product was unrelated to robotics. Mecka collects three-dimensional human-motion data by distributing iPhones and custom cameras to contributors in 12 countries, and BetaKit reports the company opened a New York office this month and has grown to 45 staff, more than half Canadian.
Financial context
BetaKit's acquisition report does not mention it, but Mecka disclosed in June that it had raised $60 million across two previously unannounced SAFE rounds, a $25 million Series A in November and a $35 million extension, led by Framework Ventures with Menlo Ventures, SV Angel, Kindred Ventures, and ex-Google DeepMind researcher Ted Xiao (now at Jeff Bezos's Project Prometheus) participating, according to Fortune. Fortune reports Mecka has raised $68 million in total since forming in 2025, projects a $100 million annual run rate from already-signed contracts, and quotes Framework Ventures cofounder Vance Spencer calling it "the fastest-growing revenue company that we've ever invested in." BetaKit separately cites a Robotics Center of Silicon Valley study putting the 2026 US robotics market at $11.4 billion, up nearly 30% year over year.
For practitioners
When evaluating third-party motion datasets or vendors, prioritize evidence of dataset provenance and audit trails, normalization and labeling schema, temporal synchronization across multi-camera captures, and contributor recruitment and diversity, the same qualities regulated-data pipelines like Docula's were built to guarantee. Mecka's growth also signals rising demand for engineers who can move data-ops discipline from regulated industries such as healthcare and finance into physical-AI pipelines.
What to watch
Whether Mecka discloses named customers or a valuation, whether competitors respond with similar tuck-in acquisitions of regulated-data teams, and how the broader physical-AI data-collection market, which also includes MicroAGI's camera-based home-cleaning data collection and Wayve's driving-camera data, consolidates as more well-funded entrants compete for the same pool of contributor data.
Key Points
- 1Mecka AI acquired three-person Vancouver startup Docula to gain regulated-data processing expertise for its physical-AI motion-data pipeline.
- 2The tuck-in follows a previously unannounced $60 million Series A led by Framework Ventures, with Mecka targeting a $100 million run rate.
- 3Buying compliance-grade data-ops talent from healthcare billing shows motion-data quality control is becoming a competitive line item for robotics-data vendors.
Scoring Rationale
This tuck-in acquisition carries more weight read alongside Mecka's previously unannounced 0 million Series A and 00 million ARR target (Fortune), showing a well-capitalized physical-AI data vendor recruiting regulated-data-pipeline expertise as it scales; still a tactical hire/acquisition rather than a new product or model release, keeping it in the solid, not major, tier.
Sources
Primary source and supporting public references used for this report.
Practice with real Ad Tech data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Ad Tech problems
