Geisel Software

A leading developer of Web apps, iOS and Android Apps, and IoT (Internet of Things) software.

[Geisel Software @Geiselsoftware]

Crops don't pause while you debug. Inside the autonomous process control software running ECO 1, the world's largest hydroponic vertical farm.

[The Geisel @thegeisel]

The crop doesn't wait for a bug fix.

Crop One built their Farm Manager software for their R&D facility in Millis, Massachusetts. Then ECO 1 came along: 300,000 square feet in Dubai, three tons of leafy greens a day, an order of magnitude larger than anything the software had been built for. Their team was onboarding new engineers and still supporting existing operations at the same time. Their VP of Digital Systems, Derek Hui, had worked with us before and brought us in.

Nine months, working directly alongside their team. We stabilized OPC-UA monitoring on modern async, and put real logging in so debugging stops being an outage. API performance improved by more than 90% on the status page and 24% on the controls page, which matters when an operator is making a real-time call based on what the interface tells them. We rebuilt and expanded the test suite, cleared security debt out of the libraries, and restructured the data architecture so the system could grow as the farm did.

Farm Manager deployed to ECO 1, and the team used it to track down PLC logic and system performance issues across the operation.

Thousands of pumps, valves, sensors, and switches, polled every 30 seconds, in one of the harshest climates on earth. A restart is not a fix when there are three tons of crops on the line.

What's the biggest jump in scale your software has had to survive.

#AgTech #VerticalFarming #IndustrialAutomation #EmbeddedSoftware #ControlSystems

[Geisel Software @Geiselsoftware]

Green tests don't mean it works. They mean it works under the conditions you held fixed.

Sun through a bay door. Glare off a recoated floor. A pallet six inches off the map. Stale data nobody age checks.

Make the test environment less cooperative before the field does.

[Geisel Software @Geiselsoftware]

Your AI prototype works perfectly. Pop the champagne. You're 12% done.

The flawless demo is step 1 of 8. Then the curve drops: edge cases, hardware failures, bugs no one can explain. Most demos stop at the bottom.

Step 1 is easy now. Step 8 is the whole job.

[Geisel Software @Geiselsoftware]

Thirty-seven seconds after ignition, Ariane 5 lost guidance. Seconds later, it was gone.
The cause: reused software encountered a condition it had never been qualified to handle.
When code controls a machine, there is no undo. Test for reality, not just success.
#EmbeddedSystems

[Geisel Software @Geiselsoftware]

700x faster, same accuracy.

A warehouse robot finds each item's grip orientation from a 3D point cloud, testing thousands of poses per item. Ported C++ to CUDA: same computations, same accuracy, only the clock changed.

#CUDA #GPUComputing

[Geisel Software @Geiselsoftware]

The readiness gap our engineers close.

[The Geisel @thegeisel]

Every robotics demo ends the same way: clean lab, robot works, everyone claps. Then someone asks, "When can we ship it?" and the room goes quiet.

Physical AI isn't a model problem. It's a deployment problem.

Benchmarks don't expose failures. Production does.

[Geisel Software @Geiselsoftware]

Everyone talks about AI making developers more productive.

Fewer people ask who trains the next generation of senior engineers when AI is doing the work they used to learn from.

That's what Tony Campbell and I discuss: https://t.co/rz9TQ6xqWf

[Geisel Software @Geiselsoftware]

AI tools are rapidly changing the software development workflow. Brian's latest video shows what that looks like in practice: one Claude Code session, from a blank screen to a live website. An interesting look at how quickly ideas can become reality.

[The Geisel @thegeisel]

Claude Code spent my money.

Not on tokens.

**It bought a domain name.**

I gave it one voice prompt and walked away.

It came up with a list of names. I picked one.

That was the last thing I did for 45 minutes.

Claude Code:

→ Registered the domain → Wrote the copy → Found the images → Got the SSL certificate → Stood up the AWS infrastructure → Deployed the site

Then one of its subagents broke halfway through.

Claude launched a replacement and kept going.

It didn’t ask me what to do. It didn’t wait for me to fix anything.

It just finished the job.

The wildest part?

I couldn’t have told you exactly what it had built on AWS until I asked it afterward to **draw me a diagram of its own architecture.**

Total damage: **2% of my weekly Claude allotment.**

I recorded the entire 45 minutes, unedited—including the parts where I’m figuring out what Claude is doing at the same time you are.

If you want to see what autonomous software development actually looks like right now, watch the whole thing:

<https://t.co/JQc7kq6mSk>

#ClaudeCode #AWS #SoftwareEngineering #AI

[Geisel Software @Geiselsoftware]

The software industry moves too fast to keep up with every headline.
Disruptive Development helps you separate what's worth your attention from what's just noise, so you can make better technology decisions.
Subscribe: https://t.co/d3QCWpqVjb

[Geisel Software @Geiselsoftware]

https://t.co/ICVByx7GAp

[Geisel Software @Geiselsoftware]

95% of enterprise AI initiatives delivered no measurable business value.

The models weren't the problem.
MIT's Project NANDA found most failures happened during deployment and integration, not model development.

Brian Geisel explains why.
👇 Link

[Geisel Software @Geiselsoftware]

Perception gets the demos. The decision layer gets the postmortems.

Real autonomy has to do three things at once: decide fast, stay safe under uncertainty, and explain its decisions. Each is solvable. Doing all three in the same control loop is where deployments stall.

[Geisel Software @Geiselsoftware]

13 years in robotics and the most exciting breakthroughs are still ahead.
At Geisel, we’ve built:
✅ Mars + warehouse simulations
✅ Computer vision systems
✅ Synthetic data pipelines
Robotics is where software, physics, and imagination collide.
What excites you most?

[Geisel Software @Geiselsoftware]

Developers don’t love engineering because they enjoy typing.
They love solving hard problems and building elegant systems. 🧠
Automate the boilerplate, not the thinking, judgment, debugging, or ownership.
AI should protect the craft, not replace it. 🛠️
#SoftwareEngineering #AI

[Geisel Software @Geiselsoftware]

AI agents need a blast radius.
Give them:
🔒 Least-privilege access
📁 Clear repo boundaries
🧪 Sandboxed execution
⏸️ Approval gates for irreversible actions
🧾 Full decision logs
Autonomy without containment is a production risk.
#AIAgents #DevSecOps

[Geisel Software @Geiselsoftware]

Autonomy can’t depend on the cloud.
Robots and drones must still decide when networks are slow, jammed, or unavailable.
Geisel Software builds on-device systems for path planning, behavior, recovery, and real-time autonomy.
#Robotics #EdgeAI

[Geisel Software @Geiselsoftware]

Vibe coding is fast. But speed without security creates risk.
Treat AI-generated code as untrustedAutomate scans early
⚠️ Be cautious with auth, data, and infrastructure
Use AI to move faster without compromising trust.
#AI #SoftwareSecurity #VibeCoding

[Geisel Software @Geiselsoftware]

Physical AI isn’t just AI connected to a machine.
Latency matters
Model accuracy isn’t enough
Safety must exist beyond the model
In the physical world, success depends on the entire system, not just the AI.
#PhysicalAI #Robotics #SystemsEngineering

[Geisel Software @Geiselsoftware]

Why are software engineers so passionate about what they do?
We push technology forward
We turn complex problems into working solutions
We build systems that create real-world impact
Watch the video to see what motivates us every day.
#SoftwareEngineering #Innovation

[Geisel Software @Geiselsoftware]

AI may increase output while destroying flow.
Measure more than generation speed:
Code produced
Delivery time
Cost to review, fix, and maintain it
Productivity isn’t how much code enters the repo. It’s how much sound thinking reaches production.
#AI #SoftwareEngineering