Skip to content

Learning, change and AI

Aksel Normann W. Edvardsen

I help people succeed at something new.

  • Through Ikomm Academy, we helped around 20 people from the welfare system into permanent jobs in IT.
  • All 13 of our IT apprentices passed their trade examination with distinction. More importantly, they grew as people and became independent, curious and ambitious young professionals.
  • We set the direction for product work at Ikomm and landed our first big commercial success built on our own software.

We did that together, and the secret was quite simple: we talked often, and people wanted each other to do well.

I have a master's in work and organisational psychology, but experience is what has shown me how much good follow-up matters. I have been building digital products with AI since the summer of 2025. Now I am looking forward to being part of a team again.

I have quite a few strings to my bow, and I am looking for an employer who sees the possibilities in that. Please take this page as an invitation.

I'm looking for someone who can see a little beyond what's in this particular video!

The year in green squares

Screenshot of the GitHub contribution graph for 2026: 2,162 contributions, nearly all from March onwards.

The year in green squares

Screenshot of the GitHub contribution graph for 2026: 2,162 contributions, nearly all from March onwards.

2,162 GitHub contributions in 2026

My GitHub contribution graph for 2026, up to 23 September. The green squares start roughly when I took a chance on Claude Code. Most of it is Mr. World Pool and the tools around it.

What I can contribute

I am happiest when something new needs to come together, whether it is an apprentice on their first day or an idea that needs to become a product. Much of the job is explaining things and clearing the way, and being there when people get stuck. It is most fun when a lot of hungry, capable people want the same thing at the same time and everyone does their part.

Apprentice leadership and learning

We built up the apprenticeship programme, which I led, and took on four or five new apprentices a year. All 13 IT apprentices passed with distinction and topped the wellbeing surveys along the way. Now they have 50 years of happy customers, colleagues and employers ahead of them.

Ikomm Academy was under way before I joined, and I was given a central role in shaping and developing it. Together we helped around 20 people from the welfare system into permanent jobs in IT. That was thanks to many good partners: professional mentors in the company, people at NAV who made room for us and cleared the way, and partner companies and course providers who worked hard so that as many as possible could succeed.

Team leadership

The apprentices were spread across several departments and offices, so the conversations every other week were perhaps the most important part of the recipe: we got to know each other, and I caught what was off, in time. In the Academy, a lot of us wanted the participants to do well, and the participants wanted the same for each other. Internships and permanent jobs were few, but from day one we made it clear that they had to help and support each other, because it was going to be brutally hard otherwise. And it worked.

Organisational development

I have been lucky to be part of a wide range of work: onboarding new employees, product organisation, reorganisations, culture work, introducing innovation systems and training people in OKRs. In 2019 I started an AI group in the company and led it. I also built our website and wrote most of its content.

In my master's I specialised in innovation and creativity, and in what encourages them. That has been a common thread in much of what I have done, in organisational work, as a team leader and in product work, and it is something I would happily work more with again. As much because my personal traits suit it as because of the theory, I think.

Product work

In 2024 I was acting product lead, as Ikomm began building commercial products on its integration platform, Ikomm Connect. We brought more product thinking into the company, and a start page product became our first big commercial success built on our own software. Today it brings in significant revenue. The prototype was mostly built by our first two IT developer apprentices, over a summer when everyone else was on holiday. There is more about that in the album.

Product work is about combining vision with execution and real needs, and about processes that work with you rather than against you. It is not easy, and we were nowhere near landing everything while I wore that hat. There is a little more about that under the things that did not go to plan.

AI and automation

Before AI became useful, I had a long no-code phase, with websites in web builders, databases in Airtable and automations in Zapier. At Ikomm I made a recruitment tool with Microsoft Forms, Lists and Power Automate that filtered and categorised the many applications for our apprenticeships, and an onboarding portal in Coda that I built from scratch. The portal was a big success, both with the apprentices and with the interns who came through. The story of the two who wanted to say hello to the directors is in the album. Throughout 2026 I have built digital services with AI and automated much of the work behind them, such as data collection, quality checks and analysis. I explain what they do under "Things I have made" further down. I like being able to show a solution while we are still discussing the problem.

Words from people who know me

Written by people I have worked with, and by some who know me from other parts of life.

Vi har kjent hverandre siden studietiden og har hatt et godt og svært innholdsrikt vennskap siden den gang. Aksel er en reflektert, allsidig og genuin person som alltid har hatt en meget god evne til å se muligheter og tenke kreativt. Jeg kjenner få andre mennesker som viser så stor nysgjerrighet i forhold til et bredt spekter av temaer-enten vi snakker om litteratur, fotball eller ai.

Harald Carlsen

Studiekamerat · Lærer

Written September 23, 2026

Aksel er skarp, kreativ, analytisk og med originale fremgangsmåter. Det som skiller ham ut, er evnen til å tenke utenfor boksen. Med Aksel får man definitivt ikke en A4-person, men en som er kapabel til å gjøre en forskjell.

Sindre Trana

Skole / Studiekamerat

The relationship was described by the page ownerWritten September 24, 2026

Hadde gleden av å jobbe med Aksel flere år i Ikomm. Han er utrolig smart og en skikkelig god humørspreder. Han har dratt i gang alt fra hopprenn og til hackatons og spillkvelder. Hans beste kvalitet er at han alltid utfordrer vedtatte sannheter, og de er det mange av innen IT-drift. Aksel satte utvikling, AI og nytekning på kartet og er veldig god til å prate med teknikere og ikke-teknikere

Robin Heby

Kollega i Ikomm 2019-2025

Written September 23, 2026

Aksel tok meg med inn i et univers av hestegalskap der vi sammen lagde prestasjonsratinger på alle galoppløp i Skandinavia, HongKong og Dubai - ja vi prøvde også oss på Mauritius 🇲🇺 Aksel har en enorm motor og stayerevne der han fortsetter der andre gir seg, og er ikke fremmed for å ta i et tak av de ekstreme variantene, og skygger ikke unna drittjobben når det er det som kreves

Gard Edvardsen

Full Galopp AS

Written September 23, 2026

Stories from my life

Moments I carry with me, from work and from the rest of life. Some I am proud of, others are mostly fun to tell now.

Or jump straight to a story

and 15 more

From my references

Excerpts from written references I received from Ikomm and NAV Lillehammer. The English excerpts below are translations of the Norwegian originals. You can open the full references under each excerpt.

He has a unique ability to motivate people, help those who are struggling get back on their feet and create experiences of mastery. At the same time, he also sets clear expectations.
NAV Lillehammer, reference dated 20 August 2025. Signed by Jens Ole Sætha, Erlend Sørnes and Jan Tore Stø. Translated from Norwegian.

See the full reference

NAV Lillehammer, reference dated 20 August 2025. Signed by Jens Ole Sætha, Erlend Sørnes and Jan Tore Stø. Translated from Norwegian.

The reference from NAV Lillehammer, 20 August 2025. Signatures redacted.
Aksel approaches tasks thoroughly and delivers solid results. Particularly worth highlighting is his ability to support the candidates in the Academy programme and our regular apprentices. Aksel has a special ability to build relationships, which in turn leads to remarkable results.
Berit Barlund, Director of HR and Organisational Development at Ikomm. Excerpt from the employment reference, September 2025. Translated from Norwegian.

See the full reference

Berit Barlund, Director of HR and Organisational Development at Ikomm. Excerpt from the employment reference, September 2025. Translated from Norwegian.

The employment reference from Ikomm, two pages. Date of birth and signature redacted.

Things I have made

Mr. World Pool is the main project, and it has taken at least 80% of my time in 2026. The rest are side projects. Some belong to the same universe, others are small detours to learn something or try out an idea. Whether they become anything more remains to be seen.

1 of 11

2025–2026 · Main project · product, data and development

Mr. World Pool

A platform that turns large amounts of horse racing data into form, ratings and analysis.

Mr. World Pool’s dark form card, showing Almaqam’s race history, ratings and performance trend.

Mr. World Pool

Mr. World Pool’s dark form card, showing Almaqam’s race history, ratings and performance trend.
More images (4)
Mr. World Pool’s form card in light mode: Almaqam’s race history, ratings and form trend.

Mr. World Pool

Mr. World Pool’s form card in light mode: Almaqam’s race history, ratings and form trend.
Mr. World Pool’s comparison view with expected ratings and projected margins between runners.

Mr. World Pool

Mr. World Pool’s comparison view with expected ratings and projected margins between runners.
Expert view in light mode: Almaqam’s form with pace, equipment and running style.

Mr. World Pool

Expert view in light mode: Almaqam’s form with pace, equipment and running style.
Comparing runners and projected distances in Mr. World Pool’s dark theme.

Mr. World Pool

Comparing runners and projected distances in Mr. World Pool’s dark theme.
Read about the work

Why

My brother and I had spent thousands of hours collecting and assessing performances over the years. I wanted to automate the manual work and build the tools I was missing as an analyst.

How

With AI as a development partner, I have built the platform and the systems behind it: data collection, quality checks, ratings and form. Underneath it all is a single rating scale, one number for how good each horse has shown itself to be, built on around 250,000 races in 17 countries, so a horse in Hong Kong can be measured against one at Ascot. Day-to-day operations now run almost entirely on their own. Building the product was demanding, but reaching people in a market that has done roughly the same thing for fifty years is harder. So much of the work is about showing and explaining, and the course and the competitions are part of that.

September 2026 · latest project · Idea, design and development

Jennys teppe

A colour planner for the patchwork blanket Mum wanted to knit for Jenny.

A knitted preview of the Mario pattern in Jennys teppe.

Jennys teppe

A knitted preview of the Mario pattern in Jennys teppe.
Read about the work

Why

My mum, who is 67, wanted to try the colours before knitting a patchwork blanket for her granddaughter Jenny. Less than fifteen minutes after she told me about the idea, the first version was online. It was more than she had imagined.

How

A small tool for colouring squares, trying colour combinations, viewing a knitted preview and saving the pattern as an image. Seeing her face when she tried the first version sparked another idea: helping small businesses with digital tasks they lack the time or skills to tackle themselves. The freelance offering is still taking shape.

Built with

  • HTML
  • CSS
  • JavaScript
  • AI

2026 · Side project · this page

BetterCV

CVs tell us too little. I have experienced that from both sides of the table.

The photo album on BetterCV’s demo profile: an open spread with a photograph, a story and textured paper.

BetterCV

The photo album on BetterCV’s demo profile: an open spread with a photograph, a story and textured paper.
Read about the work

Why

It can be hard to get through to employers who don't know you, the way recruitment works today. And when I was hiring, I would have loved to find something like this about the candidates, with real stories, a bit of personality and examples of work.

How

An experiment built with AI. The idea is that other people write about you through a link, and you choose what is shown. I'm trying it out on myself while I look for my next workplace, so the page you are reading was made with it. Many of the most exciting jobs don't quite fit an ad with a predefined role, at least not in the kind of companies I would like to work for. So the page is just as much an invitation to see possibilities. Working title: BetterCV.

Built with

  • Next.js
  • React
  • TypeScript
  • Supabase

2026 · Side project · editorial help for writers

Karlove

I wanted a good editor who could help me become a better writer.

Karlove

Read about the work

Why

This year I am writing a series of novels. It is what I do when I am supposed to be relaxing, and I find it very exciting. I do the writing itself without AI, but I needed someone who could read closely, be honest and explain why a text works or doesn't.

How

Karlove is built on a Claude skill I have filled with a large collection of notes from books on writing. It is as much a writing coach as an editor. It explains why something works or doesn't and points to its sources, so the advice rests on more than a chatbot's gut feeling. It never touches the text itself.

2026 · Side project · a safe Claude Code trial

SunRoulette

Show me where the sun is. Or just choose somewhere for me.

SunRoulette suggests Cannes, with a destination photo and sunshine, temperature and rain forecasts for the coming week.

SunRoulette

SunRoulette suggests Cannes, with a destination photo and sunshine, temperature and rain forecasts for the coming week.
Read about the work

Why

I was tired of looking up one destination after another to find the sun. I wanted to choose from places with good weather, or not have to choose at all.

How

Built in three days with Claude Code, as practice before I dared use the tool on Mr. World Pool. The app matches forecasts with destinations and lets chance make the final choice.

2026 · Side project · a course within Mr. World Pool

Do Your Homework

A free, interactive course in handicapping, built around Mr. World Pool.

Do Your Homework in light mode, with the chapter sidebar, progress and The Rating lesson.

Do Your Homework

Do Your Homework in light mode, with the chapter sidebar, progress and The Rating lesson.
Read about the work

Why

Racing has long struggled to bring in new people, because it takes time to understand. I wanted to build the way in that I wish had existed, and reach people with the mindset Mr. World Pool asks for: don't be lazy, do your homework.

How

A six-chapter introduction and a more detailed sixteen-chapter course. Learners work with real racecards, compare horses and test their judgement in exercises along the way. I built the content, illustrations and interactive platform. This screenshot is from the development version; the course is not open to everyone yet.

2025–2026 · Side project · an odds board within Mr. World Pool

Project Benter

A model that sets its own odds on horse races, and an odds board where I show it at work.

Project Benter: a Southwell odds board showing estimated win probabilities, fair odds and market comparisons.

Project Benter

Project Benter: a Southwell odds board showing estimated win probabilities, fair odds and market comparisons.
Read about the work

Why

Building a model that beats the betting market in horse racing is, according to everyone who knows anything about it, close to impossible these days. That is part of what motivates me. The model is the work I will never quite finish.

How

The model builds on the ratings and data from Mr. World Pool. It estimates each horse's chance of winning and turns that into its own odds, which I can hold up against the market. Project Benter is how I show it: an odds board and live streams where I go through races and show the model's prices along the way. It is also practice at something I am less comfortable with: stepping forward and explaining.

2026 · Side project · competitions within Mr. World Pool

#TippingComp

Free competitions for the biggest races, with a leaderboard.

#TippingComp: the Royal Ascot 2026 leaderboard with podium positions, player names and scores.

#TippingComp

#TippingComp: the Royal Ascot 2026 leaderboard with podium positions, player names and scores.
Read about the work

Why

For a couple of years I ran competitions on Twitter around the Dubai races, and it was great fun. With AI I could make it bigger, and make racing fun to follow, both for people who don't bet and in the races they haven't bet on.

How

A platform with sign-up, entries, automatic scoring and a leaderboard. The first competition was Royal Ascot in June 2026. The competitions are also a way to make Mr. World Pool known.

2025–2026 · Side project · book

The Long Edge

A practical handbook for people who want to take betting seriously.

The Long Edge: "The most boring book about betting you'll ever read".

The Long Edge

The Long Edge: "The most boring book about betting you'll ever read".
Read about the work

Why

I wanted to gather what I have learned about value, bankroll, variance and psychology into one book others could use. It is also a way to build an audience around my work.

How

About 100 pages, written to be the one book someone reads on the subject rather than the first of fifteen. No stories or anecdotes, on purpose.

2026 · Side project · tool

Bet Tracker

The spreadsheet I had used for years, prepared for other people.

The Long Edge Bet Tracker: a spreadsheet for the long game, with a chart of the results.

Bet Tracker

The Long Edge Bet Tracker: a spreadsheet for the long game, with a chart of the results.
Read about the work

Why

I had developed and used the spreadsheet for years and thought it might be useful to others. I wanted to try turning a personal working tool into a product.

How

A Google Sheets system that tracks Kelly staking, CLV, calibration and return. The point is to measure what actually tells you something over time, such as whether you beat the closing odds, not just wins and losses. I spent a couple of days getting it ready for sale, though I don't expect a huge market for it. Learning to sell has turned out to be harder than making the product.

Built with

  • Google Sheets

2026 · Side project · prototype

Textrodate

A dating platform with a get-to-know-you game, for better conversations than "wdyd".

The Textrodate front page: "Someone worth writing to".

Textrodate

The Textrodate front page: "Someone worth writing to".
Read about the work

Why

I think dating apps work poorly, especially for introverts. I liked the idea of combining a digital dating platform with a kind of get-to-know-you game that sparks slightly better conversations.

How

This is the project that has come the shortest way, and I am not sure I will take it any further. It was mostly play with an idea I liked personally, built out of curiosity with AI as a development partner.

Built with

  • Next.js
  • React
  • TypeScript
  • PostgreSQL
  • Drizzle

Experience and education

Ikomm AS

Change Adviser · February 2019–September 2025

Developing and running Ikomm Academy in partnership with NAV, establishing and leading the apprenticeship programme, organisational development and product work. Acting product lead in 2024. After I left, I was engaged to see the apprentices all the way through.

Independent projects

Since 2025

Digital products built with AI, with Mr. World Pool as the main project. Alongside it, I am writing a series of novels.

NTNU

Master’s degree in Psychology, specialising in Work and Organisational Psychology · 2016

My thesis explored creativity and cultures of collaboration, taking Pixar as its starting point.

Lillehammer University College

Bachelor’s degree in psychology · 2008–2011, completed on schedule

Alongside my degrees, I have taken individual courses and studied online, including philosophy and the history of ideas. The full list of courses from Coursera and similar platforms is on LinkedIn, though I have probably learned just as much from the courses where I only picked out what I needed.

Things that did not go to plan

Things I would do differently today, and what I took with me.

The pub quiz and product-market fit

From 2021Backwards

I love pub quizzes. In 2021, during my no-code phase before AI, I started building pubquizen.no: thousands of curated and classified questions in Airtable, with fun facts, extra background and hints for the quizmaster, and a web application made in Softr. The product was a new quiz every week, 40 questions I put together the way a good quizmaster thinks: a balance of categories and difficulty, fun facts and explanations, and enough of the questions you can reason your way to, because those are the ones that make the best team moments. Everyone who subscribed got the same questions and ran their quiz from them.

I put the project aside a few times along the way, and it took a while before the first paying customer came along. It stayed at that one, and they were very happy: municipal employees in the Trysil area who ran quizzes one autumn. They paid 500 kroner a month after I gave them half price, and I learned that I love giving service out of the ordinary. But when I did the sums, I saw it could never pay. The big quiz scenes have their own quizmasters, the chains make their own quizzes, and I would have needed customers in almost every town. Selling was hard, and putting together the perfect quiz every week was a lot of work. All credit to the quizmasters out there.

What I did afterwards

I know the theory: test the market before you build. Alberto Savoia and pretotyping are something I look up to, and here I did most of it backwards. But not everything you do is about maximising sales. It matters to me to solve problems I care about, and the pub quiz was one of them. When the sums didn't add up, it was fine to stop. Annie Duke's book Quit is about exactly that.

Mr. World Pool was also built from the inside out, but there the opportunities are international and huge, in Britain, Australia, Hong Kong and Japan. It can work if I prove myself a thousand times over, with books, content, courses, competitions and a community, with quality and by listening to the people who use it. I will never give up on Mr. World Pool. Selling it, on the other hand, I may give up on one day.

Bringing two worlds together

2024–2025Overeager

As acting product lead, I was supposed to bring more product thinking into an organisation whose whole DNA is operations, projects and sales. It taught me how demanding it is to bring together two worlds that perhaps don't quite belong together. With the Academy and the apprentices I had been given a lot of freedom, and the results there were good. Product work was different. The developers had long been somewhat in the shadow, and several people had tried before me.

I knew there would have to be some fighting, and perhaps it was also necessary for someone to dare raise their voice now and then. But I didn't quite find the balance. I was a bit overexcited, or overeager, and much of what I would do differently today is about getting results even faster and with less friction.

What I did afterwards

Things that seemed obvious to me were not as obvious to everyone else, and I didn't understand that well enough. I can still get burned by that. Now that I am selling a product to a fairly unusual niche, I am meeting the same challenge again.

But that is more about communication, about persuading people and pressing the right buttons, than about anything being wrong with the message in the first place. Niklas Luhmann, one of the maddest academics I know of, put it like this: "Communication is improbable." He may be right. But wrestling with improbabilities is, after all, one of the things I enjoy most, so I'll keep trying, again and again.

The feel-good story that wasn't

Academy, second roundBoiled over

Around the second round of Ikomm Academy, the local paper heard about the programme and wanted a feel-good story. The journalist came to visit, and several participants, who were in a vulnerable situation, had to be persuaded to take part. It was a good day. The course was nearing its end, and many had already been offered jobs. The journalist asked about one participant who was a bit desperate, because nothing concrete had come through for her yet.

I had praised her to the journalist. When the story came out, the headline read, translated from Norwegian: “Everyone got a job, but not the ‘toughest’”. A couple of paragraphs were unfortunate too, including one saying her Norwegian was weaker. At roughly the same moment as I was reading it for the first time, she called me in despair and asked if I could please help her sort it out. I have probably never been angrier in a work setting, and the journalist felt it. After a couple of rounds back and forth, his solution was to give me full editing access: "Here, write whatever you want!" And I did. There were no more stories with the local paper, let's put it that way. Luckily others came to visit later, with somewhat better routines.

What I did afterwards

Since then I have been more sceptical of journalists, and I always ask to check quotes. With E24, NRK and the Confederation of Norwegian Enterprise (NHO), it has gone fine.

And the woman in question, well, she got a job in IT in the end, she too, after one of the bravest and most impressive efforts in the history of the course. She came in with almost no prior knowledge and learned everything from scratch in six tough months, where most people would have given up on day one. And yes, the thought had certainly crossed her mind.

The model that was too difficult

Until 2025Too early

After reading about what legends like Bill Benter had achieved, I tried several times to follow in his footsteps. First with a small team of very capable colleagues, but they soon realised it would take thousands of hours, with very little chance of success. Then I hired a Romanian professor through Fiverr. It took a few weeks before I realised he probably did not have what it took, although he gave it a brave try, including a Python script that generated the heaviest Excel file I have ever opened. The laptop I used back then still shudders at the thought of it.

In my eagerness I even bought an optimisation formula from a Chinese man who said he had worked in one of the big computer-betting teams in Hong Kong. It cost several thousand kroner, and I had no idea what to do with it.

What I did afterwards

In the summer of 2025 I realised AI had made it possible, and picked it back up. Suddenly the formula was useful after all.

Steve Jobs put it like this to the students at Stanford: "You have to trust in something: your gut, destiny, life, karma, whatever. Because believing that the dots will connect down the road will give you the confidence to follow your heart, even when it leads you off the well-worn path. And that will make all the difference."

That is how I think about the job hunt too. Everything I have been doing these past few years is suddenly going to land in a company where it fits perfectly. I just don't know which one yet.

What others have measured

Ratings from Uber, where I am a passenger, and FINN, where I buy and sell. Small encounters with people who don't know me.

Uber

Passenger rating

4,95

My rating as an Uber passenger.

Screenshot · Uber

The screenshot was taken by the person who owns this page.

Screenshot from the Uber app showing a passenger rating of 4.95.

Finn.no

7 ratings

10 out of 10

On FINN since 2007, verified with BankID. Buyers and sellers mention clear communication and straightforward transactions.

“Quick and straightforward transaction, with good, trusting communication throughout.” Translated from Norwegian. Christian, seller, 30 June 2023.

Screenshot · Finn.no

The screenshot was taken by the person who owns this page.

Screenshot of the FINN profile: verified with BankID, on FINN since 2007, 10 out of 10 based on 7 ratings.

What else I am into

Books I have especially valued in recent years, films and series I am always happy to watch again, and a few posts that say something about how I think. If you want more recommendations, there is plenty more on Goodreads and IMDb.

Books I have especially valued in recent years

More recommendations:Goodreads

  • Cover of Creativity, Inc. by Ed Catmull.

    Creativity, Inc.

    Ed Catmull

  • Cover of My Struggle by Karl Ove Knausgård.

    My Struggle

    Karl Ove Knausgård

  • Cover of How to Fly a Horse by Kevin Ashton.

    How to Fly a Horse

    Kevin Ashton

  • Cover of Thinking, Fast and Slow by Daniel Kahneman.

    Thinking, Fast and Slow

    Daniel Kahneman

  • Cover of The Black Swan by Nassim Nicholas Taleb.

    The Black Swan

    Nassim Nicholas Taleb

  • Cover of Naive. Super by Erlend Loe.

    Naive. Super

    Erlend Loe

  • Cover of Maskiner som tenker (Machines That Think) by Inga Strømke.

    Maskiner som tenker

    Inga Strømke · 2022

  • Cover of Antifragile by Nassim Nicholas Taleb.

    Antifragile

    Nassim Nicholas Taleb

  • Cover of The Possibility of an Island by Michel Houellebecq.

    The Possibility of an Island

    Michel Houellebecq

  • Cover of Stoner by John Williams.

    Stoner

    John Williams

  • Cover of The Map and the Territory by Michel Houellebecq.

    The Map and the Territory

    Michel Houellebecq

  • Cover of Shark Drunk by Morten A. Strøksnes.

    Shark Drunk

    Morten A. Strøksnes

  • Cover of The Idiot by Fyodor Dostoevsky.

    The Idiot

    Fyodor Dostoevsky

  • Cover of Steppenwolf by Hermann Hesse.

    Steppenwolf

    Hermann Hesse

  • Cover of The Little Prince by Antoine de Saint-Exupéry.

    The Little Prince

    Antoine de Saint-Exupéry

Films and series I am always happy to watch again

More recommendations:IMDb

LinkedIn

Da jeg startet i Ikomm for 6,5 år siden, lovet jeg sjefen å aldri helt lære meg bedriftskulturen. Jeg skulle være annerledesmannen. Blyantspisseren i en verden av solide skruer og muttere. Og i denne planlagte mismatchen har vi skapt magi: - Vi bygget Ikomm Academy fra null til et prosjekt statsministeren ville se. - Vi lagde det jeg mener er Norges beste lærlingprogram for IT-lærlinger – uten en krone i budsjett. - Jeg rakk å brekke en overarm og ryke en sene i dueller mot økonomisjefen (som jeg vant). Nå har jeg tatt spranget videre. For selv når ting er bra, kjenner man noen ganger at det er på tide å "slippe keeperen og angripe med begge henda". Utforske litt hva mer dette livet har å by på. Derfor sa jeg opp. Flyttet fra byen. Og de siste månedene har jeg brukt all min tid på et eget prosjekt, der jeg setter det råeste av AI og meg selv på prøve. Utrolig gøy og lærerikt, og akkurat det jeg trengte nå. Jeg har skrivd litt mer utfyllende om alt sammen her: https://lnkd.in/d2HEFcn7 Tusen takk til alle kjære kolleger og samarbeidspartnere for masse gode opplevelser, samtaler og lærdommer. Jeg håper å se så mange som mulig av dere igjen, kanskje på en litteraturfestival, eller to? Eller hva med i Fotball-VM neste år? 😁 Jeg vet ikke nøyaktig hva neste kapittel bringer enda, men jeg er alltid åpen for en god prat med andre "blyanter". Ta gjerne kontakt!

LinkedIn

Vi må snakke litt om triggerhappy menn. Du kjenner typen. Utålmodig. Dominerende. Impulsiv. Selvsikker. Han som tar avgjørelser på mandags morgen før du har rukket å svelge kaffen. Og ofte så funker det. I oppgangstider er konsekvensen av å gjøre ingenting ofte verre enn å gjøre dårlige ting. Men det finnes en bakside. For ti år siden skrev jeg en masteroppgave om høykreative team med tittelen «Most Steps Wins». Hovedpoenget var enkelt: Ekte kvalitet krever flere runder tilbake til tegnebrettet enn det som er vanlig. Ingen snarveier. Forskningen viste derimot at grupper ofte klamrer seg til den første ideen som virker «god nok», og at det skal overraskende lite til for å nå dette minimumsnivået. Og hvis maktbalansen er skjev i utgangspunktet? Da er det HIPPO, "The Highest Paid Person's Opinion", som dominerer. Ikkje bra. NRKs dokumentar om Oljefondet demonstrerte et viktig poeng her. Nicolai Tangen virker på mange måter som selve arketypen av den triggerhappy mannen – utålmodig, dominerende, impulsiv, selvsikker. Heldigvis utstyrt med gode instinkter, det utgjør et hav av forskjell. Men den kanskje største forskjellen? Han virker å ha utviklet motkrefter til sin egen iboende tendens. Selv om urinstinktet ganske tydelig var frustrasjon i møte med kritikk, hadde han skapt en kultur der folk turte å si sin mening allikevel. Så stod de litt i ubehaget. Før han innrømmet, gjerne bare to minutter etter, at «fy f***, det var bra dere stoppet meg der, det hadde jo gått helt galt». Det er for meg helt glimrende ledelse. Og trolig alt som kreves for å få det beste ut av triggerhappy menn. For vi er jo definitivt mange som bør lære litt av disse myteomspunne skikkelsene, også.

anedvardsen.no

10 bøker som forandret meg. Noen bøker leser du og legger bort. Andre flytter inn. Her er de ti bøkene som har hatt størst innvirkning på den jeg har blitt.

LinkedIn · 18 September 2026

Det siste året har jeg nærmest hoppet ut av senga for å jobbe med problemer som jeg brenner for. Det var mye bra ved min forrige jobb, men når det kom til bruken av KI fikk jeg virkelig kjenne på to verdener som møttes. Ambisjonsnivået lå lysår unna mulighetene jeg så. Vi snakket mer enn vi gjorde, det var begrensninger rundt hver eneste sving, og ingen lærte noe. Til slutt tenkte jeg: Vet du hva, dette har jeg ikke tid til. Så jeg sa opp. Jeg ville utnytte dette vinduet med supersubsidierte abonnementer og rivende utvikling. Se hvor langt jeg kunne ta det hvis jeg fikk jobbe helt uhemmet. Siden har jeg virkelig fråtset i AI. Rundt 50 timer i uka med nesten vulgær produktivitet. Rett frem, full pøs. I år har jeg over 2000 commits på GitHub, i prosjekter som ville vært utopi uten AI. Jeg bygger mine egne arbeidsverktøy, og har laget en forskningsassistent som er koblet direkte på databasen som formulerer og tester hypoteser. På den måten graver vi frem den ene skjulte mekanismen etter den andre i markeder som er kjent for å være notorisk vanskelige å slå. I forrige uke, i et euforisk øyeblikk på tredemølla, fikk jeg en idé til en storstilt video som kanskje kunne hjelpe meg i markedsføringen: en episk musikksatt tale, med masse klipp og parodiske poenger som kun kan settes pris på av de innvidde. Jaggu klarte vi det også. Med mer enn 80 klipp og bilder som fades inn og ut akkurat der de skal. Enda en barriere brutt. Gjett om det gir vann på mølla. I helgen skal den selge arvesølv for meg på Finn, mens jeg koser meg på hyttetur. For en som er mer glad i gode ideer og resultater enn i å skulle gjøre ting på gamlemåten, er dette en fantastisk tid å være nysgjerrig i. Det krever fortsatt hardt arbeid og masse menneskelige vurderinger å lage bra ting, mye mer enn det 37,5 timer i en vanlig norsk bedrift noensinne krevde av meg. Og jeg var ikke kjent for å hvile på laurbærene, der heller. Nå kjenner jeg en voksende utålmodighet med alle de skråsikre KI-meningene som florerer om dagen, og som ser ut til å bygge på lite faktisk bruk. Her er det innlegg både for og imot som utmerker seg. Når du faktisk jobber med dette, bruker det hver dag, gjør noe mer enn bare å dippe tåa i vannet med noen banale oppgaver du løser med ChatGPT, lærer du også om begrensningene, og hva vi mennesker fortsatt gjør bedre. Som for eksempel salg. Jeg trekker tilbake alt jeg har sagt om selgere i hele mitt liv. Seneca skrev at en gammel mann kan ha alderen som eneste bevis på sitt lange liv. Jeg har tenkt mye på den setningen. Siden jeg sa opp, har jeg blitt ett år eldre. Men det føles som jeg har fått tjue år mer erfaring. Jeg er glad og litt stolt over meg selv for at jeg turte. For alt jeg har lært, og for at jeg fortsatt er så sulten på mer. Dette er energi og erfaring jeg tar med meg når jeg nå stepper opp i jakten på mine neste kolleger, mitt neste sted å bo og de neste problemene som skal få meg opp om morran. Jeg skal selvfølgelig bruke AI til det også.

Get in touch

Perhaps the role doesn't exist yet. If so, I would be glad to talk about what it could look like, and to show you a couple of projects.

Read all the way down here?

Maybe I wasn't what you were looking for. If you still have a piece of advice about the page, what should go, be said differently or be added, I'd be glad to have it. It only goes to me.

So this is advice, not an anonymous message.

What should go, be said differently, or be added? Between 20 and 3000 characters.

Goes only to Aksel. Never shown on the page.