Ali Rathore

Trusting agents with enterprise data

Fri, 01 May 2026 00:00:00 +0000

An execution model for autonomous agents that survives enterprise security review.

Retrieval you can audit

Wed, 01 Apr 2026 00:00:00 +0000

Document intelligence where every answer carries its provenance.

Fifteen services, one install

Sun, 01 Mar 2026 00:00:00 +0000

Delivering a fifteen-service data platform into customer Kubernetes environments in one step.

A lockless hash table in a database core

Wed, 01 Jan 2014 00:00:00 +0000

Inside SAP HANA, where a single data structure tripled parallel query throughput.

Robots that made burgers

Fri, 01 Jan 2016 00:00:00 +0000

Directing the software for autonomous burger assembly at Momentum Machines.

The honest default is no

Wed, 24 Jun 2026 00:00:00 +0000

Four systems that do nothing alike share their most important line of code, the one that runs when the system has no idea, and in all four it refuses.

Inducing the schema instead of supplying it

Mon, 22 Jun 2026 00:00:00 +0000

The unsolved part of document AI is discovering the columns, not filling them, and a column should only exist if it pays for itself in bits.

I built it one validation too early

Sun, 21 Jun 2026 00:00:00 +0000

The abstractions I have had to delete were not wrong, they were premature, built to answer a question the work had not yet asked.

A good average hides the column that matters

Sat, 20 Jun 2026 00:00:00 +0000

A synthetic dataset that scores well on average can still be useless and unsafe, so the only scores worth trusting gate on the worst part and fail closed.

Point, don't drag

Thu, 18 Jun 2026 00:00:00 +0000

Drag-and-drop is the wrong handle for steering a model that builds your interface, because the thing you grab and the thing it wrote stopped being one-to-one the moment a list appeared.

The harness remembers what the agent forgets

Wed, 17 Jun 2026 00:00:00 +0000

The coding agent starts every build with no memory of the last one, so the system that runs it got better by writing down where the agent kept failing.

Letting an agent merge to main

Mon, 15 Jun 2026 00:00:00 +0000

Autonomy is not something the model has. It is the set of gates you are willing to put around it, and an audit trail of everything it did inside them.

Watching an agent work

Sat, 13 Jun 2026 00:00:00 +0000

The interesting part of an agent is not that it acts, it is the shared room where you can see what it sees, point at things, and take the controls back.

A turn is not a request

Fri, 12 Jun 2026 00:00:00 +0000

An agent’s unit of work is a long-lived process, and the moment you treat it like an HTTP request you lose the run every time a laptop sleeps.

Managing engineers who never sleep

Tue, 09 Jun 2026 00:00:00 +0000

A year of running AI coding agents turned into an accidental management apparatus, written one rule at a time.

What it takes to let agents touch enterprise data

Mon, 08 Jun 2026 00:00:00 +0000

Trust in agentic systems is an architecture property, not a model property.

Your gold set is lying to you

Mon, 25 May 2026 00:00:00 +0000

Reference answers are artifacts with bugs, and they are the only software in the stack that nobody code-reviews.

An answer you cannot audit is worth nothing

Mon, 18 May 2026 00:00:00 +0000

Provenance is a data-model property, not a UI feature, and it cannot be retrofitted.

Authorization for answers

Mon, 27 Apr 2026 00:00:00 +0000

We have learned to secure what agents do. The harder question is what they are allowed to know.

Evaluating systems that answer in sentences

Mon, 20 Apr 2026 00:00:00 +0000

Evaluation infrastructure is the difference between a demo and a product.

LLM spend is an attribution problem

Mon, 06 Apr 2026 00:00:00 +0000

A month of anonymous model spend taught me that cost governance for AI is identity infrastructure, and the discipline that fixes it already has a name.

Data infrastructure and AI infrastructure are different disciplines

Mon, 30 Mar 2026 00:00:00 +0000

They optimize for different things, fail in different ways, and the hard problems live at the seam.

Why your knowledge graph isn't helping your RAG

Mon, 09 Mar 2026 00:00:00 +0000

We nearly published the fashionable verdict that GraphRAG is hype. The audit found something better, the structural reasons the published architectures break on enterprise data.

The physical world does not retry idempotently

Mon, 02 Mar 2026 00:00:00 +0000

What burger-making robots taught me about agents acting on production data.

The model designs, the code enforces

Mon, 16 Feb 2026 00:00:00 +0000

In an LLM ingestion pipeline, the model gets the judgment that requires reading the document and nothing else. The surprise is how badly the prompt wants to violate that in both directions.

Database lessons for the agent era

Mon, 09 Feb 2026 00:00:00 +0000

Agent platforms are rediscovering, one incident at a time, what database engines settled decades ago.

Tables are images

Mon, 26 Jan 2026 00:00:00 +0000

Meaning in a table lives in its geometry, and the architecture that survives a real corpus is a per-table router, not an ideology.

What regulation does to architecture

Mon, 12 Jan 2026 00:00:00 +0000

Years of healthcare engineering taught me that compliance, taken seriously, is a design input that produces better systems.

When documents become databases

Mon, 15 Dec 2025 00:00:00 +0000

Schema discovery is a corpus-statistics problem before it is a modeling problem.

Free like a puppy, times fifteen

Mon, 17 Nov 2025 00:00:00 +0000

A product assembled from open source is not the sum of its components. It is the resolution of their disagreements.

Forcing the race

Mon, 20 Oct 2025 00:00:00 +0000

Race conditions are not hard to test. They are untested, because teams accept probabilistic reproduction.

About

Mon, 01 Jan 0001 00:00:00 +0000

I work at the convergence of data infrastructure and AI, as a founding engineer on an agentic data fabric platform for enterprises with complex operational data estates. My scope runs from API contracts and identity models through agent execution, retrieval, evaluation, and Kubernetes delivery, plus the hiring and the standards that keep all of it coherent.

The arc that got me here is unusual on purpose. I started inside the SAP HANA database core, building lockless data structures and chasing nanoseconds with SIMD and RDMA. Then a robotics startup, directing the software that assembled burgers autonomously. Then a big-data platform at SparklineData that Oracle acquired, and Spark-as-a-Service for Fortune 500 workloads at Oracle itself. Then engineering leadership in healthcare: telehealth at eVisit, health-equity fintech at Soda Health. In between, I founded Lakeshore Labs, an AI agents consultancy serving enterprises across Southeast Asia and the Gulf.

Resume

Mon, 01 Jan 0001 00:00:00 +0000