Blog

    Engineering Notes on Agent-First Software

    How AltSlate Labs builds agent-first software and the compact models behind it: what we tried, what we measured, and what did not work.

    All articles

    Engineering

    An AI That Grades Its Own Code Keeps Its Own Misreadings

    We trained a model on answers from running its own code. A misreading in that code never got fixed. What did fix it, and what that fix then broke.

    Nov 6, 20268 min read
    Engineering

    Fine at 32 Bits, Broken at 4: Where the Softmax Bug Shows Up

    We mapped where a known training bug shows up. In 32-bit training it barely matters. In 8- and 4-bit it hits most of every batch, attention first.

    Oct 31, 20267 min read
    Engineering

    A New Fix for a Training Bug Is Real. The Accuracy Gain Mostly Isn't.

    We re-ran a softmax loss fix in PyTorch across 480 training runs. The bug is real, the accuracy gain small. Where 16-bit maths happens mattered more.

    Oct 26, 202610 min read
    Engineering

    A 16‑Line Code Check Beat Automated Prompt Tuning

    Every rewritten prompt made our small-model chatbot worse, automatic or not. A 16-line check that refuses incomplete records fixed it. Five lessons.

    Oct 24, 20268 min read
    Engineering

    RL Does Exactly What You Reward. The Environment Decides What That Is.

    We built an RL environment for oil-well troubleshooting. RL optimised exactly what we rewarded, loopholes included. The worked examples did the rest. Four lessons.

    Oct 15, 20269 min read
    Engineering

    AI Confidence Scores Depend on the Options You List

    Adding an option, even one nobody picks, moved a decision model's scores. Longer lists moved them more. What we measured and five practical lessons.

    Oct 14, 20268 min read
    Engineering

    Fewer, Closer Examples Beat a Smarter Lookup

    We tried making AI lookups push look-alike examples away. It failed. Reading fewer, closer examples worked better. Three practical lessons for noisy data.

    Oct 10, 20267 min read
    Engineering

    Image Checks From a Few Dozen Labelled Photos

    Calibrated yes/no, choice and score answers about images from a few dozen labelled photos, with an abstain flag. Coarse checks hit 93–95%; fine detail fails.

    Oct 9, 202611 min read
    Engineering

    Speed or Rules? What Caching a Decision Model Costs

    Storing options made our decision model 5× faster on big menus but worse at stated rules. Training restored it on synthetic rules; real ones are unproven.

    Oct 7, 20268 min read
    Engineering

    Agent Harness: How We Run 40 AI Agents in Production

    What an agent harness is and how ours runs about forty AI agents in production: one graph, permission gates, typed tools, human approvals and evals.

    Oct 1, 20269 min read
    Product

    The Assistant That Asks Before It Acts

    An operations assistant where AI agents do the legwork and changes to the business arrive as previews a person approves. What it is like to use.

    Sep 25, 20266 min read
    Engineering

    Half the Memory, None of the Answers Lost

    Small AI models get expensive when they must remember a long history. One fix flopped. The other halved that memory for free, without changing an answer.

    Sep 19, 20269 min read
    Engineering

    The Verifier's Blind Spot

    We built a "self-improving" LLM cascade to get cheaper every round. It got quietly worse — and every dashboard said it was fine. What we measured, and why you cannot see it from inside.

    Sep 4, 20268 min read
    Market

    Technology Finally Reached the Delivery Layer

    Cloud rebuilt the front door of services firms and left the workshop untouched. Agents walk into the workshop — and that changes the economics of professional services.

    Aug 14, 20268 min read
    Product

    Why We Build Agent‑First, Not Agent‑Bolted‑On

    Bolting an LLM onto an existing app gets you a demo. Designing the agent as the operating core — with humans in command — gets you a product. Here is how we think about the difference.

    Aug 12, 20269 min read
    Engineering

    Serverless That Actually Scales (and Stays Cheap)

    Serverless is not automatically cheap or scalable — it is a set of trade-offs. Here are the patterns we lean on to get fan-out compute that bills to zero when idle.

    Jul 28, 20269 min read
    Product

    Production, Not Pilots — How We Build

    Most AI work stalls at the demo. We design, build, and operate production systems end to end — applying AI where it earns its place and conventional engineering everywhere else.

    Apr 15, 20266 min read