Coram AI Pitch Deck: All 6 Slides + Teardown

Coram AI, a North American cybersecurity startup, raised $13.8M in a 2024 Series A round using a concise 6-slide deck. The presentation positions the company as a provider of 'AI Agents for the Physical World,' focusing on the massive $120B labor market for security and operations professionals. The deck emphasizes technical scalability, claiming to process 100,000 hours of video daily using on-premise Large-Vision and Language Models (LVLMs). While the deck is visually polished and highlights a strong customer testimonial from North Cross School, it notably omits a detailed team slide, speci…

Key takeaways

• The deck positions the product as 'AI Agents' rather than just a security camera platform to tap into the $120B labor market (Slide 5).
• Coram AI claims to process 100,000 hours of video daily, which it equates to nearly half of YouTube's daily uploads (Slide 8).
• The technology stack utilizes on-premise LVLMs (Large-Vision and Language Models) including InternVL2-8B and OpenCLIP (Slide 8).
• A customer testimonial from Zack DeMoss at North Cross School highlights the deployment of 150 cameras on the Coram system (Slide 3).
• The market opportunity is framed by 500 million security cameras existing outside of China (Slide 5).
• The deck uses a highly condensed 6-slide format, skipping traditional elements like a dedicated team slide or detailed roadmap (Slides 1-14).
• The product is designed to work with 'Any IP Camera,' lowering the barrier to entry for existing facilities (Slide 8).
• The closing slide asserts 'Rapid growth' and 'Very capital efficient' status, though no specific revenue figures are provided in this version of the deck (Slide 14).

Frequently asked questions

How much did Coram AI raise with this deck?

As reported by Business Insider, Coram AI raised $13.8M in a Series A round in 2024. The deck itself identifies the round as 'Series A' on the title slide but does not specify the dollar amount or valuation.

What is the core technology behind Coram AI?

According to Slide 8, the company uses an on-premise 'Coram Point' server that runs Large-Vision and Language Models (LVLMs) like InternVL2-8B and OpenCLIP. This allows for high-volume video processing (100,000 hours daily) while maintaining low-bandwidth cloud connectivity.

Who are the target customers for Coram AI?

The deck shows applications in schools (North Cross School), warehouses, and retail environments. Slide 3 specifically highlights 'Fast Incident Resolution' and 'Operational Insights' like counting pallets shipped, suggesting a dual use for security and business operations.

Does the deck include a team slide?

No. While Slide 14 mentions a 'Proven team for the mission,' there is no slide in this 6-page set that lists the founders, their backgrounds, or previous company exits.

What is the 'labor' angle mentioned in the market slide?

Slide 5 argues that while $100B is spent on physical security hardware and software, a larger $120B is spent on the 1.5M security and operations professionals in the US. Coram AI aims to capture this larger segment by using AI agents to replace human labor.

Coram AI pitch deck: the facts

Company

Coram AI

Year

2024

Stage

Series A

Slides

6

Sector

AI, Cybersecurity

Deck type

Series A Pitch Deck

Outcome

$13.8M Raised

Headquarters

North America

Coram AI pitch deck PDF

The full Coram AI deck is embedded on this page and can be read slide by slide in the browser — no download or account required. Each slide is covered in the breakdown above.

What the Coram AI pitch deck was used for

This deck is Coram AI’s 6‑slide **Series A** fundraising presentation used for its $13.8M round announced in January 2025 (often described as raised in the prior year), focused on AI-native physical security.[2][5][9][11] It pitches Coram’s CL:RAM platform, which uses multimodal large language models (LVLMs) to turn existing IP cameras into autonomous AI agents that process 100,000 hours of video per day on‑premise.[2] The deck frames the company’s mission as bringing AI agents to the physical world by modernizing video security in sectors like schools, hospitals, warehouses, manufacturing, and logistics.[2][5] The round was led by Battery Ventures with participation from 8VC and Mosaic Ventures, with proceeds earmarked for expanding product lines and growing go‑to‑market and engineering teams.[2][5][10][11]

Business model: Coram AI provides an AI-native, hardware-agnostic video security platform that layers generative AI and multimodal large language models on top of existing security cameras and systems to detect threats (e.g., firearms, slip-and-fall incidents) and enable fast video search and investigation.[1][2][5][10]

Round

Series A[2][5][9][13]

Raised

$13.8M Series A[2][5][9][10][11][13]

Lead investor

Battery Ventures[2][5][9][11][13]

Investors

Battery Ventures, 8VC, Mosaic Ventures

Founded

2022[2][5]

Founders

Ashesh Jain, Peter Ondruska[5][10][15]

Headquarters

Sunnyvale, California, United States[5][7][13]

Industry

AI-powered physical security / video security software[2][5][10]

Year: 2024–2025 (round announced mid-January 2025, with some coverage describing the capital as raised in 2024).[2][5][9][10][11]

Total funding: Approximately $30M total funding as of the $13.8M Series A announcement; later sources report $66M total after a subsequent $35M Series B, indicating significant additional funding after this deck’s round.[1][3][4][6][7]

Use of funds as presented: Coram intended to use the Series A funds to build more product lines and grow its go‑to‑market and engineering teams for its AI-native, hardware-agnostic video security platform.[5][10]

What happened after the Coram AI deck

The Series A deck helped Coram AI secure $13.8M in funding led by Battery Ventures, enabling the company to expand product lines and go‑to‑market efforts for its CL:RAM AI security platform.[2][5][10] Following this round, Coram reported substantial growth in revenue and customer deployments, which supported a subsequent $35M Series B that increased total funding to around $66M and scaled the plat

• Closed a $13.8M Series A round led by Battery Ventures with participation from 8VC and Mosaic Ventures, bringing total f
• Battery Ventures partner Marcus Ryu joined Coram AI’s board alongside the Series A financing.[5]
• Subsequent reporting indicates that after the Series A, Coram’s revenue grew approximately fourfold and its customer bas
• By the time of the Series B, Coram’s platform was reportedly deployed at more than 1,500 sites across North America, inc
• In June 2026, Coram announced a $35M Series B co‑led by Ansa Capital and Battery Ventures, with participation from UP.Pa

What the Coram AI deck got right

• Demonstrated large-scale LVLM video processing capability. The CL:RAM architecture is explicitly "architected for massive scale" with on‑prem multimodal LVLM inference, processing 100,000 hours of video daily—nearly half of YouTube’s daily uploads—demonstrating a clear claim of substantial real-world throughput that is unusual in early-stage AI decks.[2]
• Hardware-agnostic integration with existing IP cameras. The deck emphasizes that Coram works with "Any IP Camera" and operates in "Low-bandwidth" environments, positioning the product as hardware-agnostic and easy to layer onto existing infrastructure, which is compelling for cost-sensitive physical security buyers.[2][5]
• Modern, multimodal AI architecture with named components. The architecture slide names modern components such as InternVL2‑8B, OpenCLIP, and vector databases, signaling a contemporary, differentiated multimodal AI stack rather than generic computer vision, which aligns with the claim of "Differentiated technology."[2]
• Crisp articulation of vision, team, market, and financial discipline. The reasons-to-be-excited slide clearly articulates a **big vision**—AI agents for the physical world—while also highlighting a "Proven team," "Massive market," "Very capital efficient," and "Rapid growth," condensing key investor hooks into a single, easily digestible summary.[2]
• Clear real-world use cases and early customer adoption. External reporting around this Series A round cites real customers in sectors such as K‑12 education and retail and later references over 1,500 deployed sites, reinforcing that the deck’s focus on practical deployments and safety use cases is grounded in actual adoption.[5][6][10]
• Strong investor syndicate and governance backing. The round attracted a high-quality lead in Battery Ventures and repeat participation from 8VC and Mosaic Ventures, with a seasoned operator (Marcus Ryu) joining the board, which underscores investor confidence in the team and the opportunity.[2][5][9][11]

What could have been stronger

• Limited quantified proof of model accuracy and reliability. The deck touts processing "100,000 hours of video processed daily" and comparisons to YouTube’s uploads, but available text does not show benchmarks about model accuracy (e.g., false positive/negative rates, latency, or reliability in real-world deployments), leaving investors guessing about core performance risk for safety-critical use cases.[2]
• Growth and efficiency claims lack detailed supporting metrics in visible slides. While the deck asserts "Rapid growth" and "Very capital efficient," the excerpted slides do not provide concrete metrics such as customer count, annual recurring revenue, deployment numbers, or unit economics at the time of the Series A, making it harder to assess traction and efficiency from the deck alone.[2]
• Insufficient explicit treatment of privacy, compliance, and security controls. The technical slide mentions components like InternVL2‑8B, OpenCLIP, vector DBs, and on‑prem LVLMs, but the available text does not explicitly explain how Coram manages data privacy, security, and regulatory compliance (especially for schools and healthcare), which are critical concerns in video surveillance.[2][5]
• Integration and deployment complexity not fully addressed in excerpted slides. The architecture slide references "Any IP Camera" and "Low-bandwidth" operation, yet the visible material does not delve into integration complexity, retrofit costs, or limitations with legacy hardware, leaving some implementation and deployment risks underexplored.[2][5]

How an investor would read this deck

What draws attention

• Coram AI is targeting a **massive physical security market** using AI-native software that works with existing cameras, which can be attractive because it avoids hardware replacement and leverages already-installed infrastructure.[2][5][6]
• The CL:RAM platform claims **100,000 hours of video processed daily** via on‑prem multimodal LVLMs, suggesting meaningful scale and offering investors a chance to back a real-world, production AI system rather than a purely experimental model.[2]
• The Series A was led by **Battery Ventures** with participation from **8VC** and **Mosaic Ventures**, and later rounds add investors like Ansa Capital and UP.Partners, indicating strong venture backing and signaling quality.[2][3][4][5][7][12]
• Founders **Ashesh Jain** and **Peter Ondruska** have prior experience in AI and autonomy, and the company already counts customers such as K‑12 school districts and retail chains by the time of its Series A, showing early commercial traction.[5][10]
• Reported post‑Series A outcomes (4x revenue growth and tripled customer base before the Series B) suggest that the growth narrative in the deck was at least partly validated.[3][6]

Risks that stand out

• Video-based AI for physical security is highly sensitive to model errors; misdetections could lead to missed threats or frequent false alarms, potentially eroding customer trust and creating liability risk.[1][2][5]
• Operating large multimodal LLMs on-premise at the scale of 100,000 hours of video daily may carry significant infrastructure, maintenance, and optimization challenges that could affect margins and performance.[2][5]
• The company operates in regulated and privacy-sensitive environments (schools, healthcare, logistics), where evolving legal standards around surveillance, biometrics, and data retention could constrain product capabilities or require costly compliance work.[2][5]
• Competition is intensifying in AI-powered physical security and video analytics, including from incumbents adding AI features and startups building similar camera-agnostic platforms, raising the bar for differentiation and sales execution.[3][5][6]
• Coram’s growth story involves rapid scaling across more than a thousand sites by the time of its Series B; sustaining deployment quality, support, and model performance at scale is operationally demanding and may expose execution risk.[3][4][6][7]

Questions this deck invites

• What were Coram AI’s concrete traction metrics at the time of this Series A deck (e.g., number of deployed sites, paying customers, annual recurring revenue, and deployment growth over the prior 12 months)?[2][5]
• For CL:RAM’s LVLM-based video analysis, what are the documented accuracy, false positive/negative rates, and latency across key use cases like firearm detection and slip-and-fall alerts, and how do these compare to human monitoring or compe
• How does Coram handle data privacy, retention, and regulatory compliance (FERPA, HIPAA, GDPR/CCPA where applicable) when processing continuous video streams from sensitive environments like schools and hospitals?[2][5]
• What are the total cost of ownership and integration steps for retrofitting "Any IP Camera" with Coram’s platform, and how have customers evaluated ROI in terms of reduced incidents, faster investigations, or labor savings?[2][5]
• Who are Coram’s main competitors in AI-native physical security and video analytics, and what proprietary advantages (models, data, integrations, or workflows) sustain the "Differentiated technology" claim over time?[2][5][6]

What founders can take from the Coram AI deck

• A concise Series A deck can still be effective if it combines a **big vision** (AI agents for the physical world) with one or two highly concrete proof points, such as processing 100,000 hours of video per day and integrating with any IP camera.[2]
• Positioning AI software as a **layer on existing hardware** ("Any IP Camera," low bandwidth, on‑prem LVLMs) can strengthen the value proposition by lowering adoption friction and avoiding capital-intensive hardware rollouts.[2][5]
• Clearly articulating reasons to be excited—big vision, differentiated technology, proven team, massive market, capital efficiency, rapid growth—helps investors quickly understand why this opportunity could be outsized, even before detailed numbers are presented.[2]
• Using **named model components** (e.g., InternVL2‑8B, OpenCLIP, vector DBs) signals technical depth and modern architecture, which can build credibility with technically savvy investors while still being framed in business outcomes like safety alerts and fast search.[2]
• Tying the deck’s claims to specific customer segments (schools, hospitals, warehouses, manufacturing, logistics) and use cases (firearm detection, slip-and-fall detection) makes an abstract AI vision tangible and easier for investors to map to market size.[2][5][10]
• Follow-on coverage showing 4x revenue growth and tripled customer base after the Series A illustrates how a strong alignment between deck narrative (rapid growth, capital efficiency) and execution can later reinforce investor confidence and support larger rounds.[3][4][6]

Coram AI pitch deck: common questions

What does Coram AI do?

Coram AI is an AI-native video security company that uses generative AI and multimodal large language models to convert existing IP cameras and security infrastructure into autonomous agents for physical security, enabling capabilities like firearm detection, slip-and-fall alerts, and fast video search.[1][2][5][10]

How much did Coram AI raise in the Series A associated with this deck, and when?

Coram AI raised a **$13.8 million Series A** round announced around January 16–19, 2025, with some coverage describing the capital as raised in 2024.[2][5][9][10][11] The deck you are researching is the 6‑slide Series A pitch used for that fundraise, emphasizing CL:RAM’s ability to process 100,000 hours of video daily and replace manual monitoring with autonomous agents.[2]

Who invested in Coram AI’s Series A round?

The **Series A** was led by **Battery Ventures**, with participation from existing investors **8VC** and **Mosaic Ventures**.[2][5][9][10][11][13] Battery partner Marcus Ryu, co‑founder and former CEO of Guidewire Software, joined Coram’s board as part of the financing.[5]

What were the main messages in Coram AI’s Series A pitch deck, and did the company deliver on them?

Coram AI’s Series A deck positioned the company as building AI agents for the physical world on top of existing security cameras, highlighting differentiated LVLM-based technology, a proven team, a massive physical security market, capital efficiency, and rapid growth.[2] Subsequent reporting states that since closing this $13.8M Series A, Coram’s revenue grew roughly fourfold and its customer base tripled ahead of its later $35M Series B.[3][6]

How did Coram AI plan to use the Series A funds raised with this deck?

The Series A financing was intended to help Coram build more product lines and expand its go‑to‑market and engineering teams.[5][10] The deck also implicitly positioned the funding as fuel to scale CL:RAM’s LVLM inference infrastructure, support more deployments across sectors like K‑12 education and logistics, and deepen its AI-agent capabilities for physical security.[2][5]

Sources

Funding and outcome facts on this page were researched on 2026-08-30 from the pages below.

• Raised a $13.8M Series A
• Deck: Coram AI Raises $13.8M to Bring AI Agents in ...
• Coram Raised $35 Million to Turn Security Cameras Into AI Detectives
• We've Raised $35M to Build Autonomous Agents That ...
• Coram AI Raises $13.8M Series A
• Coram Raises $35M to Turn Cameras into AI Detectives
• Coram AI raised $35 million in Series B funding ...
• Coram AI: Funding, Team & Investors

Coram AI pitch deck slides

What each slide of the Coram AI pitch deck says

Slide 1

1 — — CL:RAM Ne - TOS Al Agents For The Physical World ¥ pil § SBS Cem TY & . Series A p——— Mad

Slide 3

CliRAM 3 We Protect and Optimize Every Physical Space BS ara: = EE == "i . =A Me EE Find person sing at 3pm the acy room 5 455 pallets shipped today ® Fast Incident Resolution Enhanced Physical Security Operational Insights “We are buying that 1-minute of extra time that can save lives.” Zack DeMoss, North Cross School, 150 cameras on Coram

Slide 4

Cil:RAM 5 Massive Market With Software Replacing Labor Physical security Labor hardware & software $20B Cameras 500M $100B a $120B security cameras spent on SW & replace labor spent on security outside of China HW for physical and operations security professionals* $38B Alarms ; = *1.5M security and

Slide 5

Cl:RAM Architected for Massive Scale LVLM* Inference 100,000 hours of video processed daily with on-prem LVLMs, equivalent to nearly half of YouTube's daily uploads Coram Point Secure Cloud End Customer InternVL2-8B OpenCLIP Any IP Camera M Vector DB Low-bandwidth *LVLM are Multi-modal LLMs they are abbreviation of Large-vision and Language models On-prem LVLM

Slide 6

CL:RAM Reasons To Be Excited About Coram ° Big vision: Al agents for the physical world ° Differentiated technology @ Proven team for the mission @ Massive market @ Very capital efficient @© Rapid growth

Slide text above is read directly from the Coram AI deck PDF embedded on this page.

Related fundraising guides (24)

• CAMP4 Therapeutics Pitch Deck Teardown: A $100M Masterclass in Biotech Platform Validation
• Innovafeed Pitch Deck Teardown: A $250M Masterclass in Industrial Scale and Strategic Partnerships
• Chptr Pitch Deck Teardown: Modernizing the $120B Death Industry for Digital Natives
• Martian Pitch Deck Teardown: A $9M Seed Masterclass in LLM Orchestration and Interpretability
• Chattermill Pitch Deck Teardown: A Masterclass in Enterprise Social Proof and Expansion Strategy
• micro1 Pitch Deck Teardown: Leveraging Growth Momentum and AI Vetting for Global Engineering
• Checkstep Pitch Deck Teardown: A 7-Slide Product Showcase for AI Content Moderation
• Viso Pitch Deck Teardown: A 42-Slide Masterclass in Computer Vision Infrastructure
• Claira Pitch Deck Teardown: Solving the Legal Logic Gap in Financial Contracts
• Vespa.ai Pitch Deck Teardown: A Corporate Spin-Out Strategy for High-Scale Vector Search
• CMR Surgical Pitch Deck Teardown: A $600M Masterclass in Ecosystem-Led Medical Robotics
• Yelp Pitch Deck Teardown: A Post-IPO Growth Story Focused on Local Ad Dominance
• Code First Girls Pitch Deck Teardown: A Masterclass in B2B Social Impact Scaling
• RegioHelden Pitch Deck Teardown: A Masterclass in Local Marketing Unit Economics
• Cowboy Pitch Deck Teardown: A Masterclass in High-End Lifestyle Branding and Urban Mobility
• Curio.io Pitch Deck Teardown: Selling High-Engagement Audio Journalism
• Craft.io Pitch Deck Teardown: A $8M Series A Masterclass in Product-Led Narrative
• Thematic Pitch Deck Teardown: Leveraging Creator Networks for Scalable Music Promotion
• Cube Pitch Deck Teardown: A $30M Series B Masterclass in Narrative Simplicity
• Staffjoy Pitch Deck Teardown: A Masterclass in Algorithmic Workforce Automation
• Domino Data Lab Pitch Deck Teardown: A Masterclass in Enterprise Social Proof
• Limetree Pitch Deck Teardown: A Case Study in Early-Stage Family Media Preservation
• EcoLocked Pitch Deck Teardown: Decarbonizing Construction via Biochar Admixes
• Cove.Tool Pitch Deck Teardown: A Masterclass in Visualizing Efficiency Gains

This deck's categories (1)

• Series A & Series B Pitch Deck Examples

Decks from the same year (1)

• 2024 pitch deck examples

Decks with a similar raise (1)

• Pitch decks that raised $10M to $50M

Browse companies alphabetically (1)

• Pitch decks from companies starting with C

Decks in the same category (12)

• MySQL 2003 Series B Pitch Deck Breakdown
• Oscar Health Pitch Deck Breakdown
• LinkedIn 2004 Pitch Deck Breakdown: How Networks Became the Next Big Thing
• DoggyHut Pitch Deck Breakdown: The Perfect Series A Deck For A Company That Does Not Exist
• Monzo Pitch Deck Breakdown: Building the Bank of the Future
• Pendo Series B Pitch Deck Breakdown
• Taster Pitch Deck Breakdown: Scaling Digital Restaurant Brands via a Licensing Model
• Zio Studios Pitch Deck Breakdown: A Transmedia Pivot from Services to IP
• Hadean Pitch Deck Breakdown: Proving Technical Scale Through Enterprise Case Studies
• Now Pitch Deck Breakdown: A $29M Series A Built on Municipal Bond Securitization
• Hangry Pitch Deck Breakdown: Scaling Multi-Brand Virtual Restaurants in Indonesia
• Seedtag Pitch Deck Breakdown: Leveraging the Post-Cookie Era with Contextual AI

More pitch deck teardowns (16)

• Cooleaf Pitch Deck Breakdown: A Masterclass in High-Impact, Low-Text Visual Storytelling
• Copado Pitch Deck Breakdown: Dominating the Salesforce DevOps Ecosystem
• COPL Pitch Deck Teardown: Consolidating Wyoming Assets and Frontier Discovery Potential
• Copper Cow Coffee Pitch Deck Teardown: A Masterclass in Demo Day Minimalism
• Copper Cow Coffee Pitch Deck Breakdown: A Masterclass in Visual DTC Storytelling
• Copper Cow Pitch Deck Breakdown: A Masterclass in Visual Storytelling and Retail Traction
• Copy.AI Pitch Deck Breakdown: A Masterclass in Traction-First Series A Storytelling
• Coralogix Pitch Deck Breakdown: Redefining Observability with Stateful Streaming
• Core Capital Venture Indonesia Pitch Deck Teardown: A Hybrid Real Estate and Fintech Play
• CorgiAI Pitch Deck Breakdown: The $2M Seed Deck That Buried Its Best Two Slides
• Corkt Pitch Deck Teardown: A Hyper-Local Approach to Social Photo Sharing
• CornFinger Pitch Deck Teardown: A Hyper-Local Niche Apparel Strategy
• Corti Pitch Deck Teardown: A Masterclass in High-Stakes Healthcare AI Storytelling
• Cosimo Drinks Pitch Deck Teardown: A Thesis on the Indian Sparkling Fruit Beverage Gap
• Counteract Pitch Deck Breakdown: A $42M Masterclass in Thematic Climate Fundraising
• Counterpart Pitch Deck Breakdown: A $30M Series B Masterclass in Market Education

Recently published pitch deck teardowns (12)

• Strong Roots Pitch Deck Breakdown: Scaling Frozen Plant-Based Food via Retail Traction
• StreamLoan Pitch Deck Breakdown: A Masterclass in Visual Contrast and Traction Storytelling
• Supercast Pitch Deck Breakdown: A Masterclass in Creator Economy Unit Economics
• SupportLogic Pitch Deck Breakdown: A 7-Slide Masterclass in Technical Product Narrative
• Taskly Pitch Deck Breakdown: A Minimalist Approach to Productivity and NLP
• The Plate Pitch Deck Breakdown: A 9-Slide Creator Economy Play for the Culinary World
• Twine Pitch Deck Breakdown: Solving the Creative Marketplace Liquidity Problem
• Unmind Pitch Deck Breakdown: Scaling Workplace Mental Health with a Preventative Model
• UnitesUs Pitch Deck Breakdown: Every Section An Investor Expects, And No Evidence Behind Any Of Them
• Vango Pitch Deck Breakdown: A Minimalist Approach to the $30B Wall Decor Market
• VCV Pitch Deck Breakdown: A Minimalist Approach to AI-Driven Recruitment Automation
• Verbit Pitch Deck Breakdown: Scaling Transcription with a Hybrid AI-Human Model

Fundraising library · Pitch deck examples · Investor directory · Founder database