Energy / Grid Inspection Software · Deep dive

Buzz Solutions

Palo Alto AI grid-inspection platform that closed a $20M Series A in August 2026 led by S3 Ventures — PowerAI ingests drone, helicopter, ground and fixed-camera imagery from transmission lines, distribution networks, substations and utility-scale solar and returns georeferenced defect lists in about 0.6 seconds per image, versus the 1-2 minutes utility field engineers still spend on manual review.

emerging

The question that decides it: Does PowerAI defend software margin as Skydio and Percepto bundle their own defect-detection models into the drone hardware from above, and as PG&E-style in-house computer-vision teams at Duke, ComEd, Southern Company and AEP absorb the same workflow from below — or does Buzz get squeezed into a mid-market niche of Tier-2 utilities and municipal power authorities that cannot afford to build their own vision stack?

My take ★★ Very interesting ★ Interesting − Maybe ✗ Not interesting

Snapshot

Buzz Solutions is a nine-year-old Palo Alto AI company that closed a $20M Series A on August 4, 2026 led by S3 Ventures, with GoPoint Ventures, HearstLab and Blackhorn Ventures. PowerAI ingests drone, helicopter, ground and fixed-camera imagery of transmission lines, distribution, substations and utility-scale solar, and returns georeferenced defect lists in roughly 0.6 seconds per image versus the one-to-two minutes a utility field engineer spends on manual triage. Named customers include Dominion Energy, American Electric Power and the New York Power Authority. The company reports tripled customer count and 400% revenue growth in the trailing twelve months to August 2026, on top of doubled revenue in 2024 (Latka pegs 2024 ARR at roughly $2M). Total raised is now approximately $30.8 million.

Founding story

Kaitlyn Albertoli and Vikhyat Chaudhry met at Stanford in a LaunchPad entrepreneurship course. Albertoli’s prior life ran through JPMorgan Chase and a sustainable-food nonprofit; Chaudhry had built a drone from scratch during his final year at Delhi College of Engineering to sample Delhi’s air quality, then completed a Stanford MS in civil and environmental engineering studying wind-turbine siting with drone imagery. Buzz Solutions was founded in 2017 to commercialise the wind-turbine idea. Within roughly a year — after informational interviews with alumni across the energy industry — they pivoted to the substantially larger and messier problem of electric grid inspection. Chaudhry was named to Forbes 30 Under 30 in Energy in 2021; the co-founder profile that Blackhorn published in 2023 describes the two as a classic technical-plus-commercial pairing where Albertoli owns utility sales and Chaudhry owns the vision-model roadmap.

How it works

The PowerAI pipeline is a fairly conventional utility-CV stack, well executed. Imagery arrives from three sources: drone flights (increasingly on Skydio X10 platforms and DJI hardware, sometimes autonomous), helicopter patrols using RGB and thermal cameras, and ground crews with handheld devices. Buzz automatically matches each image to the correct pole, tower or substation asset via GIS coordinates, then runs an ensemble of computer-vision models: object detectors that locate insulators, crossarms, conductors, hardware, splices and vegetation; anomaly classifiers that flag broken insulators, missing pins, corroded conductors, bent or cracked crossarms, damaged pole tops, and burn marks; and thermal models that detect hotspots on splices and equipment. Defects are ranked by severity and pushed into the utility’s work-management system as prioritised work orders, with the image evidence attached.

Buzz publicly cites 0.6-second-per-image inference latency and cites a Skydio-integrated live demo where a Dominion Energy pilot flew nine transmission structures in twenty minutes and pushed 138 images into PowerAI. PowerGUARD is the sibling product: an edge-deployed model that runs 24/7 on fixed cameras at substations, doing perimeter intrusion detection, PPE monitoring, person-down alerting, and fire/smoke/arcing detection. Southern California Edison is a named PowerGUARD user; NYPA deployed PowerGUARD through the EPRI Incubatenergy Labs program.

Product and business overview

Two named products. PowerAI is the transmission, distribution, substation and utility-scale solar visual analytics layer, sold to utility asset-management, vegetation-management and reliability teams. PowerGUARD is the substation edge product, sold to physical security and operations teams. The company has publicly said Series A capital will extend PowerAI into utility-scale solar imagery — a natural adjacency because solar farms already fly regular drone missions and defect taxonomies are well defined. Integrations targeted at utility IT are the standard set: GIS (Esri), work-order and asset-management systems (IBM Maximo, GE APM, Salesforce Field Service), and drone fleet software (Skydio Cloud, DJI FlightHub). The company markets a partner-first posture rather than replacing customer-owned drones — imagery-agnostic ingestion is the wedge into utilities that have already spent millions on airframes.

Business model and pricing

Buzz sells enterprise software subscriptions to utilities, priced per asset under management (per mile of line, per substation, per solar-farm MW) with usage tiers layered on top. Public pricing is not disclosed; a $2M 2024 ARR spread across the three named Tier-1 customers plus a handful of smaller ones implies six-figure ACVs, consistent with how eSmart Systems, Sharper Shape and Optelos price into the same buyers. PowerGUARD is a separate SKU sold per substation with a hardware-plus-software element (the edge model runs on customer-provided or Buzz-supplied compute at the site). Revenue is recognised ratably over the contract term; deals are multi-year, which is why customer count roughly tripling over twelve months produces disproportionate ARR jumps.

Traction over time

The trajectory, dated where possible, is: 2017 founding; 2020-2022 early pilots with EPRI, NYPA and a handful of utility R&D groups; 2022 seed round with HearstLab and Blackhorn; 2023 revenue roughly $1.3M (Latka); March 20, 2024 $5M growth round led by GoPoint Ventures with Blackhorn and MaC; 2024 revenue roughly $2M with 100% YoY growth and 40% team expansion (company disclosure via 2025 blog post and September 2024 Forbes profile); August 4, 2026 $20M Series A. The company’s own August 2026 language — “tripled customer count and 400% revenue growth over the past year” — implies a run rate in the high single-digit millions of ARR entering the Series A, with the three named Tier-1 utilities plus expansion into utility-scale solar as the growth vectors.

Market analysis

Third-party estimates cluster the grid-inspection software-plus-services layer at $5-10B, inside a broader utility drone-inspection TAM most analysts put at $3B in 2026 growing to $9B by 2034 (Intel Market Research, 14% CAGR). Verified Market Reports pegs the wider power-inspection-drone market including hardware at $11.6B in 2026 to $34.3B by 2034 (12.1% CAGR). Three structural forces: (1) US wildfire liability — PG&E’s post-2018 bankruptcy remains the reference case — is forcing western utilities toward more frequent inspections; (2) FAA BVLOS rulemaking, culminating in PG&E’s 2026 approval, is unlocking economics that make drone-first inspection cheaper than helicopter patrol; (3) grid modernization and data-centre load growth are pushing utilities to catch up on decades of deferred capex documentation.

Competitive intel

Six rivals, each attacking a different edge. Skydio owns the drone hardware and increasingly the on-drone analytics; Buzz is one of many partners in Skydio’s integrations catalogue — comfortable until Skydio decides defect detection is a first-party product. Percepto relaunched its Inspection Intelligence Platform in June-July 2026 as a vertically integrated hardware-plus-AI stack — the exact bundle a utility CIO wants from one vendor. eSmart Systems has the deepest reference base in Europe and North America and overlaps PowerAI feature-for-feature. Sharper Shape (~$21.2M raised) is smaller but has a Living Digital Twin narrative and DSP-management partnerships. Optelos, LiveEO and Overstory attack the vegetation-management adjacency; LiveEO and Overstory use cheaper satellite imagery to cover territory drones cannot. And the utility itself — PG&E’s 180-drone program, ComEd’s Advanced Image Analytics, Duke and Southern Company’s internal builds — is the largest and most under-reported threat. PrecisionHawk’s December 2023 Chapter 11 after ~$120M raised is the cautionary reference for pure-play third-party inspection.

History and evolution

2017 Stanford founding as a wind-turbine analytics idea; ~2018 pivot to grid inspection; 2019-2021 early utility pilots through EPRI; 2020-2021 NYPA collaboration through Incubatenergy Labs deploying PowerGUARD at substations; 2021 Chaudhry named to Forbes 30 Under 30 in Energy; 2022 seed round with HearstLab and Blackhorn; 2023 Southern California Edison PowerGUARD deployment; March 20, 2024 $5M GoPoint-led growth round; September 2024 Forbes profile framing Buzz as wildfire-and-blackout prevention; 2024 100% YoY revenue growth to ~$2M ARR; PowerAI expansion into distribution and utility-scale solar during 2025; August 4, 2026 $20M Series A led by S3 Ventures. No public reporting of major crises, executive turnover or product recalls — for a nine-year-old venture-backed company, either a genuinely clean run or carefully managed press.

What people say

The case for. Named Tier-1 utility references — Dominion Energy, American Electric Power, New York Power Authority, Southern California Edison — are unusually strong for a company at Buzz’s revenue scale. Skydio’s public integrations catalogue lists Buzz as a partner, giving the product a distribution channel into 280+ utility customers. Blackhorn’s 2023 founder profile and NVIDIA’s blog coverage both frame the company as commercially credible rather than pilot-stuck. Forbes’ September 2024 profile positioned Buzz as one of the more concrete wildfire-prevention plays. Investor stickiness through three rounds (Blackhorn, HearstLab, GoPoint) is a legitimate signal.

The complaints. Public reviews are thin — no meaningful G2, Capterra or TrustRadius footprint as of August 2026, which is itself a signal for a product that has been in market for seven years. Glassdoor results for “Buzz Solutions” surface a different company (a Montreal fleet-management business), meaning the Palo Alto grid company has essentially no employee-review presence to read. The 2024 ARR figure disclosed via Latka (~$2M across three customers) suggests customer concentration risk that the 2026 tripling claim only partly resolves. And the elephant in the room — the pure-play third-party inspection playbook — is exactly what PrecisionHawk failed at.

Outlook: the open question

Bull case: Buzz owns the analytics layer end-to-end for utilities that will not build in-house, and defends that layer against drone-hardware vendors moving up. Bear case: PowerAI ends up as a feature in Skydio Cloud or Percepto AIM, and Buzz is squeezed into a mid-market of municipal power authorities and Tier-2 utilities that cannot afford their own vision stack — real, but not venture-scale. The Series A funds ~24-30 months of runway assuming ~$8-10M annualised burn. The deciding data points: whether utility-scale solar produces net-new logos rather than existing-account upsell, and whether at least one hyperscaler-adjacent partnership (Palantir Foundry, Databricks lakehouse-native deployment, or an AWS/Microsoft utility bundle) materialises before Skydio or Percepto ship a first-party analytics competitor at a bundled discount.

How to attack it

A well-funded attacker has three usable wedges. First, sell the software as a training-data flywheel to utility CV teams, not as a black-box inference API. Buzz’s model is closed; PG&E, ComEd and Duke are building internal teams and would pay for a data-labelling and model-management platform (Scale AI meets Snorkel meets utility-specific defect taxonomy) that respects their desire for model ownership. Buzz’s structural weakness is that its go-to-market assumes utilities outsource inference forever; a growing set of utilities want to own the model and buy only the tooling. Second, attack the vegetation-management adjacency with satellite-plus-drone fusion. LiveEO and Overstory prove the satellite lane; a hybrid entrant combining Planet or Airbus satellite tasking with drone follow-up on hot spots collapses the cost curve versus Buzz’s drone-only workflow. Third, bundle inspection with insurance. The wildfire-liability angle Buzz uses in its marketing is stronger as an insurance product: an MGA that reads inspection cadence and defect-remediation SLAs into premium pricing for wildfire and equipment breakdown makes the software a distribution channel for the real profit pool. Buzz’s weaknesses are legible: thin public reviews, unclear pricing, no disclosed valuation, customer concentration (three named Tier-1 customers doing most of the ARR heavy lifting), single-region HQ, no visible Europe or Asia presence, dependence on Skydio’s partnership goodwill. Each is exploitable by an entrant that raises more capital, ships in Europe first, or bundles with a hardware or insurance partner.

Adjacent-segment play

The imagery-plus-defect-taxonomy stack generalises to any infrastructure owner running regular visual inspections with liability exposure to defect-driven failures. The obvious lanes: (1) Telecom towers — American Tower, Crown Castle and SBA Communications each own 30-100k towers with quarterly inspection cycles; the defect taxonomy (guy wires, antennas, mounts, corrosion) is close enough that PowerAI could be repackaged. Sitetracker and drone MSPs already play. (2) Utility-scale solar — Buzz has flagged this as their Series A expansion, but the buyer persona is different (asset owner or O&M contractor, not grid ops); Above Surveying, Raptor Maps and Zeitview compete. (3) Oil and gas midstream — pipelines, compressor stations, tanks; Percepto’s Chevron reference proves the buyer exists. (4) Rail — Class I railroads run regular track and catenary inspection; Bentley and Fugro compete but the vision layer is thin. What does NOT generalise is workflows requiring specialised sensors Buzz has not built for (LiDAR-heavy inspection, methane sensing, non-visual NDT). The most attractive adjacent bet is utility-scale solar because the motion (drone-first, defect-heavy, O&M cost pressure) is closest to what Buzz already sells — which is why Raptor Maps has already built a $30M+ business there and will not concede quietly.

Sources and further reading

• Buzz Solutions Raises $20 Million Series A Led by S3 Ventures to Power the Next Generation of Grid Intelligence — EIN Presswire, August 2026
• Buzz Solutions Raises $20M in Series A Funding — Finsmes, August 2026
• Buzz Solutions Raises $5M to Accelerate Adoption of its AI-Powered Platform — PRWeb, March 2024
• Buzz Solutions Harnesses AI To Prevent Wildfires And Blackouts — Forbes, September 2024
• Buzz Solutions Uses Vision AI to Supercharge the Electric Grid — NVIDIA Blog
• Founder Profile: Kaitlyn Albertoli, CEO & Co-founder, Buzz Solutions — Blackhorn Ventures
• Vikhyat Chaudhry, CTO, COO & Co-Founder of Buzz Solutions — Interview Series, Unite.AI
• Buzz Solutions integration profile — Skydio
• How Buzz Solutions hit $2M revenue and 3 customers in 2024 — Latka / GetLatka
• Utility Inspection Drone Market 2026 to 2034 — Intel Market Research
• Percepto Unveils AI-Powered Inspection Intelligence Platform for Energy Infrastructure — UASweekly, July 2026
• PG&E is getting closer to remote, autonomous line inspections — Latitude Media