In Parallel: Product, team and how it was built
In Parallel (AI meeting capture and shared company context for AI tools, Finland): Five enterprise customer pilots, February 2026. Evidence grade E: no revenue figure was checked. Team: 2 co-founders, Markku Mäkeläinen and Kristian Luoma; the About page lists 5 leaders; headcount not published. Launched October 2023; still running.Added 5 November 2026
- Outcome
- Five enterprise customer pilots, February 2026
- Evidence
- E no revenue figure was checked
- Launch
- October 2023
- Status
- Still running
- Team
- 2 co-founders, Markku Mäkeläinen and Kristian Luoma; the About page lists 5 leaders; headcount not published
- Outside funding
- None reported
- Main channel
- Not stated
What is not verified
No revenue figure was found. The only traction figure is "Five enterprise customer pilots deployed", from a Techarena 2026 writeup of 14 February 2026, repeated as five enterprise teams in a sponsored profile of 6 April 2026; neither says whether the pilots pay. We looked for a payment-processor profile, a founder interview with figures and Finnish financial statements, and found none. A reader could look for the company's filed accounts in the Finnish trade register or a customer case study. This story is grade E: the business is real and its product, price and customers are described from public pages, but no revenue figure was checked. Aiolum shows it as an idea worth studying, not as proof that it makes money.
In Parallel calls itself a context layer: a store of up-to-date company facts (decisions, commitments, risks) that AI assistants read before they answer, so that their answers rest on what the company decided last week and not on an old plan. A recorder joins meetings, speech is turned into text, and an AI model extracts the decisions and tasks. AI tools connect to it over MCP, the Model Context Protocol, an open standard by which an AI assistant calls an outside service for data. Each shared item carries one of three levels of detail, from the full record to the conclusion only.
In Parallel Oy was entered in the Finnish trade register on 10 October 2023, with its business ID dated 3 October 2023, and is based in Helsinki. Its About page names two co-founders, Markku Mäkeläinen (CEO) and Kristian Luoma (product), and three more leaders. At the Techarena 2026 startup event, a writeup of 14 February 2026 listed "Five enterprise customer pilots deployed".
This story has no revenue figure. The company publishes prices from €29 a user a month, and the jury writeup and a sponsored founder profile speak of five enterprise teams, but no source says how many customers pay or how much.
How much revenue does In Parallel make?
Each row shows the evidence grade and the source numbers from the list at the end of this page. The grades are explained in How Aiolum verifies.
| Fact | Value | Grade | How we know | Sources |
|---|---|---|---|---|
| Enterprise pilots, February 2026 | five | C | Techarena 2026 TOP46 writeup on UpNext, 14 February 2026: "Five enterprise customer pilots deployed." It does not say whether the pilots were paid. The same writeup reports customer results (25% faster blocker resolution, 30% less reporting work), which are the company's pitch claims, not checked. | [5] |
| Enterprise teams using it, April 2026 | five | C | Sponsored founder profile on UpNext, 6 April 2026: "the five enterprise teams already running on In Parallel". | [6] |
| Revenue | not found | E | No downloaded source gives a revenue figure, a paying-customer count or a contract value. | [5, 6] |
| Company registration | In Parallel Oy, business ID 3394334-7, dated 3 October 2023; registered 10 October 2023 | C | Finnish trade register open data (YTJ). Main business line: computer consultancy and computer facilities management. The About page says "Founded in 2023, based in Helsinki." | [7, 2] |
| Months from registration to the pilot count | 28 | calc | Calculated: October 2023 to February 2026. | [7, 5] |
| Founders | Markku Mäkeläinen (co-founder and CEO) and Kristian Luoma (co-founder, product) | C | About page. The UpNext profile says Luoma met Markku in Silicon Valley and that both worked as operators at a Swiss company that grew from 40 to 1,300 people in 18 months. | [2, 6] |
| Team size | not published; 5 leaders on the About page | C | The About page lists the two co-founders and heads of sales, design and customer success. The careers page says it is not hiring for a specific role this quarter. No source suggests more than 10 people. | [2] |
| Prices | €29 and €69 a user a month; Enterprise custom | C | Home page pricing section, read 5 November 2026. 20% less when prepaid for a year. | [1] |
| AI model and speech providers | Amazon Bedrock for language models; AssemblyAI for transcription | C | Security page, list of six sub-processors, all processing data in the EU. The language model itself is not named; the changelog speaks of "an upgraded AI model". | [3, 4] |
| Certifications | ISO 27001, ISO 42001 and SOC 2 Type II | C | Stated on the home and security pages and in the February 2026 writeup. The certificates themselves were not read. | [1, 3, 5] |
| Outside funding | none found | C | No downloaded source mentions an investor or a funding round. No source says it is self-funded either. | [2, 5, 6] |
How to read this table: grade C is what the company writes on its own site, what an event writeup reported from its pitch, or a public register entry. A "calc" value was computed on this page. No value is grade A or B, and no revenue figure was found.
| Plan | Per month, billed yearly (US$) | Per month, billed monthly (US$) | Limit |
|---|---|---|---|
| Professional | 25.90 (€23.20) | 32.37 (€29) | 120 Knowledge Units a user a day; notes, chat, pages, workspaces, AI over MCP; teams under 50 users |
| Business | 61.62 (€55.20) | 77.02 (€69) | 250 Knowledge Units a user a day; shared workspaces with access rights; up to 5 app integrations |
| Enterprise | custom | from 50 users; one annual price; unlimited integrations, SSO, a named contact |
Prices are per user a month and exclude VAT. Monthly billing is at list price and can be cancelled any month; prepaying a year saves 20%. Knowledge Units are a usage allowance pooled across the company. Every user who signs in needs a seat, including clients added to a workspace; AI agents need none, and reads from the customer's own AI tools over MCP are free within fair use. One plan applies to the whole company.
When did In Parallel launch, and how did it grow?
From a company registration in October 2023 to weekly product releases and five enterprise pilots in 2026. Each entry carries its grade.
- October 2023CIn Parallel Oy receives its business ID on 3 October and is entered in the Finnish trade register on 10 October. [7]
- 14 February 2026CNamed in the Techarena 2026 TOP46 list of standout startups, with five enterprise customer pilots and ISO 27001, ISO 42001 and SOC 2 Type II. [5]
- 6 March 2026CEarliest entry on the public changelog page: meeting reports can be edited in the product. [4]
- 6 April 2026CA sponsored founder profile of Kristian Luoma on UpNext describes the "living execution plan" and five enterprise teams using it. [6]
- June 2026CThe connector for AI tools such as Claude, Cursor, ChatGPT and Copilot becomes available in every company. [4]
- 1 November 2026CLatest weekly changelog entry; the home page sells three plans from €29 a user a month. [4, 1]
Why did In Parallel work?
Four causes, in the order they mattered. Each one is a reusable lesson.
Extract decisions, not transcripts
The company treats the transcript as disposable and keeps the decisions and commitments extracted from it. A smaller record is easier to keep current and to share with AI tools than a pile of raw meeting notes.
Permission levels on each item
Each shared record has three levels: full, sources unnamed, or conclusions only. The same decision can reach the whole company without the discussion behind it, which large companies need before they let AI tools read meeting content.
Sell trust to enterprises
EU hosting, a separate database schema per customer, no training on customer data, and ISO 27001, ISO 42001 and SOC 2 Type II. These answer the security questions an enterprise buyer asks before a pilot.
Price per person, free for agents
Seats are per user, and AI agents that read the context need no seat, so a customer can switch AI tools without changing its bill.
How it was built
These details come from the company's own pages and their page source. They are grade C. [3, 4, 1]
- Language model processing runs through Amazon Bedrock, with hosting, databases and file storage on Amazon Web Services in Frankfurt.
- Meeting recordings are transcribed by AssemblyAI.
- Stripe takes payments, Postmark sends transactional email, PostHog measures product use, and Intercom runs in-app support chat.
- An MCP server lets Claude, ChatGPT, Copilot, Cursor and Gemini CLI read the shared context with the user's permissions.
- The marketing site is built with the Astro web framework, according to its page source.
Tools named in the sources
Each one opens its page in the Tools tab, with every story that uses it.
Who tried this before In Parallel, and why did they fail?
Three companies in our database of failed startups tried to collect a team's knowledge or work in one place and lost to the large platforms. The values below are from that database and are not independently verified.
Niles
The promiseAn AI chatbot for Slack that answered questions with knowledge gathered across a company's teams.
Why it diedSlack added its own AI features, and better, faster and cheaper competitors followed.
LiveLoop
The promisePowerPoint files turned into live documents that a team could edit together.
Why it diedMicrosoft bought it in 2015 and then built the same features into Office.
Woven
The promiseAn intelligent calendar that brought a person's and a team's scheduling into one place.
Why it diedIt could not beat Google Calendar or make users pay for what they got free elsewhere.
The failed startups come from our database of 1,749 dead startups, retold here in our own words. Its entries are AI-assisted summaries of public sources, so these cards carry no evidence grade.
What changed
All three sat on top of a platform (Slack, Office, Google Calendar) whose owner added the same feature. In Parallel faces the same risk: its own comparison table puts it next to Microsoft Work IQ and Claude Projects, which do part of the job inside their own products.
What differs is that language models can now turn a meeting into a list of decisions without a person writing it, and MCP lets one store serve every AI tool a company uses. In Parallel's bet is that a company wants this record outside any one model vendor, with access rules it controls.
Can one person copy In Parallel now?
Hard for one person.
The product is simple to describe, but enterprise buyers ask for security certifications, EU hosting, single sign-on and a sales process, and this team brought decades of enterprise experience to it.
Where it is still open: Narrower versions for one function, such as account handoffs between sales and customer success, which the site lists as a use case.
- 1Pick one recurring job that managers do by hand after meetings, such as updating a plan or a status report.
- 2Keep the extracted decisions, not the transcript, and let each item carry its own access level.
- 3Expose the record over MCP so customers can use any AI tool.
- 4Get the security certificates that your buyers ask for before the first pilot.
What could go wrong for In Parallel?
The risks are the mirror image of the causes above.
What is not verified
No revenue figure was found. The only traction figure is "Five enterprise customer pilots deployed", from a Techarena 2026 writeup of 14 February 2026, repeated as five enterprise teams in a sponsored profile of 6 April 2026; neither says whether the pilots pay. We looked for a payment-processor profile, a founder interview with figures and Finnish financial statements, and found none. A reader could look for the company's filed accounts in the Finnish trade register or a customer case study.
The founder profile is sponsored content
The UpNext article of 6 April 2026 is a Flashpoint POV Spotlight; its footer says Flashpoint Global produces each spotlight with the founder in a "Future Narrative session". It is the company's own account, not independent reporting.
Meeting recording and consent
The product joins meetings and transcribes them. The security page lists GDPR compliance, EU processing and retention controls; no downloaded page says how other participants in a meeting are told that a recorder is present.
Competition and model providers
On this site it is closest to Wave AI, a meeting note-taker, and Recall, a personal knowledge base, but it sells to enterprises rather than to individuals. Its own home page compares it with Microsoft Work IQ and Claude Projects. It depends on Amazon Bedrock for language models and AssemblyAI for transcription, according to its sub-processor list.
Sources
The grade next to each source is the best grade that the source can support. "Read in full" means that the text of the page was downloaded and the numbers were checked in it.
- C In Parallel home page: The context layer for your company's AIRead in full on 5 November 2026, including the pricing section and the page source. Source for the product, the prices and the comparison table.
- C In Parallel: About usRead in full on 5 November 2026. Source for the founding year, the founders and the leadership list. Also read https://www.in-parallel.com/careers.
- C In Parallel: SecurityRead in full on 5 November 2026. Source for the sub-processors, the hosting and the certifications.
- C In Parallel changelogRead in full on 5 November 2026. Entries from 6 March 2026 to 1 November 2026.
- C UpNext, 14 February 2026: Techarena 2026: Epiminds wins, TOP46 standoutsRead the section on In Parallel. A writeup of the startup's pitch; source for the five pilots and the certifications.
- C UpNext, Dr. Paul Newton, 6 April 2026: Managers Can't Lead While They're Buried in CoordinationRead in full. A sponsored Flashpoint POV Spotlight on Kristian Luoma. Source for the founders' background and the five enterprise teams.
- C Finnish Patent and Registration Office, YTJ open data: In Parallel OyRead the JSON record on 5 November 2026. Source for the business ID, the registration date and the business line.
- C US Federal Reserve, H.10 Foreign Exchange Rates, release of 2 November 2026Read the euro row. Rate of 30 October 2026, US$1.1163 per euro, used for the price conversion.
- C Database of failed startups, local exportLocal export. AI-written summaries, not independently verified. Source for the failed predecessors.
Run this company? Send corrections or a verified figure to aiolum@livyden.com