Email me
Open to work · APAC · remote · hybrid Auckland --:--

Darshil Kothari

Site Reliability Engineer · One NZ via Infosys · Auckland · 8 years in production

Site Reliability Engineer. I take AI agents from demo to production and I test them until they break.

Production agents, evals and guardrails on enterprise systems, built on eight years of reliability engineering at an airline, two global retailers and a national telco.

tailor for
See my work→

Worked with Lufthansa · Burberry · Macy's · Alshaya · Vodafone NZ · One NZ (via IBM and Infosys)

try this first

Most portfolios describe incidents. Break this one.

Press the button and this page really slows down. An AI agent then diagnoses it from live browser measurements, shows its evidence and waits for your approval before fixing anything. That is the same loop I build in production for real customer journeys.

  1. 1You inject the fault
  2. 2The agent detects the SLO breach
  3. 3It gathers evidence and finds the cause
  4. 4You approve the fix and it heals
scroll for the story

01 · ai work

Agents in production, with evidence.

Four pieces of work from One NZ. Each one is an agent or an eval that has to hold up on enterprise systems, so each one shows its evidence and stops for a person where it matters.

A · incident triage

Agentic incident triage

Problem
One customer journey crosses six systems. When it breaks, every system alerts and people argue about where the fault is.
I built
An agent that runs grounded DQL on Dynatrace Grail and cites evidence for every claim. Each proposed fix carries a risk rating and waits at a human approval gate.
Result
A person approves each fix with the evidence in front of them. The numbers are in the results section.
PythonFastAPIDynatrace Grail · DQLMCPClaude · OpenAILangGraph

B · daily batch check

A daily agent for five batch phases

Problem
A batch run moves through five phases and each one needs checking every day.
I built
A daily agent that checks all five phases over an MCP integration and reports what it found.
Result
A consistent daily check of every phase, run by the agent.
PythonMCPTool use

C · evals as release gates

Eval harnesses that gate releases

Problem
A prompt or retrieval change can look fine in a demo and quietly make answers worse.
I built
Golden sets that measure hit@k, MRR, latency and cost per 1,000 queries, plus prompt-injection tests. A change has to pass before it ships.
Result
Quality, cost and latency are measured together on every change, so a release is decided by numbers.
PythonpytestHybrid BM25 + vectorCross-encoder rerank

D · adoption

Microsoft Copilot rollout

Problem
Teams needed to use Microsoft Copilot well in their daily work, not just have a licence for it.
I built
A rollout with adoption training across teams.
Result
Adoption training delivered across teams.
Microsoft CopilotTrainingEnablement

skills

Languages

Python, FastAPI, pytest, TypeScript, SQL, DQL

LLMs and agents

Claude and OpenAI APIs, Amazon Bedrock, LangGraph, LangChain, MCP, tool use

Retrieval

RAG with hybrid BM25 and vector search and cross-encoder rerank

Evals

Golden sets, graders and prompt-injection gates

Safety

Guardrails and least privilege

Platform

AWS, Azure, Kubernetes, OpenShift, Terraform

Observability

Dynatrace, Splunk

Customer work

Workshops and solution design

02 · the gap

Most systems work on a laptop. Far fewer ship reliably, every day, to millions of real customers. The distance between those two things is where I have spent the last eight years. I rarely remember the clever fix. What I remember is the pattern that meant nobody ever had to fix it again.

← works on a laptop chaosorder trusted in production →

03 · Lufthansa via IBM · 2018 to 2020 · seven months on site in Frankfurt

Running Lufthansa's customer marketing platforms.

In plain EnglishI kept the Adobe platforms behind Lufthansa's website and customer campaigns running and made sure only the right people could change them.

My first production job was DevOps Engineer at Lufthansa, running Adobe Experience Platform, which serves their customer-facing site and Adobe Campaign, which runs their customer campaigns. Both were business-critical and used by teams on every continent.

  • Instances, deployments, upgrades, certificates. All handled with minimal disruption to live operations.
  • Access governance. Roles and permissions across globally distributed teams and agencies, with monitoring and audit checks.
  • Under the hood: Adobe Experience Platform and Adobe Campaign instance management, deployment scheduling and upgrades across timezones and certificate lifecycle management.
  • The risk: one wrong permission could expose customer data or push a campaign live in the wrong market.
Adobe Experience PlatformAdobe CampaignAccess governanceRelease mgmtGerman 🇩🇪

04 · London & New York (remote) · Burberry, Macy's via IBM

Deployment standards for two global retailers.

In plain EnglishI made releasing new software safe and routine for two global retail brands.

At Burberry, each client team was shipping in its own way. I built the CI/CD and infrastructure automation, then set the deployment standards every pipeline had to meet. At Macy's the stakes were retail uptime, so I automated operations in Python and Bash, with monitoring that caught anomalies before shoppers did.

  • Deployment standards that made pipelines measurably more reliable across client teams.
  • Automation first for high-availability retail systems, with alerting tuned to catch anomalies early.
  • Under the hood: quality gates, image scanning, canary releases and automated rollback, rolled out across Jenkins and GitLab.
  • The outcome: defects caught where they are cheap to fix, not in front of customers. Recognised with the Burberry Shining Star.
CI/CDJenkinsPython · BashMonitoring🏅 Shining Star

05 · Remote from Mumbai · Vodafone NZ via Infosys · 2022 to 2023

Multi-cloud Kubernetes that survives a bad day.

In plain EnglishI kept a telco's systems online across two clouds and broke things on purpose to prove they could recover.

This was a telco on AWS and Azure at once. I built CI/CD pipelines with Jenkins and Docker, ran multi-cloud Kubernetes with Terraform and Ansible and did the root-cause analysis when things broke. Then I broke things deliberately, as a Gremlin-certified chaos engineer.

  • Multi-cloud K8s on AWS and Azure, fully managed as infrastructure-as-code.
  • Incident RCA with Nagios and Splunk to keep production reliable.
  • Under the hood: EC2, S3, ELB and RDS provisioning, probes, autoscaling and failover, plus Python and Bash automation versioned in GitLab.
  • What changed: failures got found on a Tuesday afternoon, not at 3am. SLOs gave teams a number to negotiate over.
KubernetesAWS · AzureTerraform · AnsibleSplunkGremlin certified

06 · Auckland · One NZ via Infosys · 2023 → now

Incident agents that find the cause and wait for a human.

In plain EnglishI build AI agents that spot outages early and find the cause. They suggest a fix and a human always approves it.

I am a Site Reliability Engineer at One NZ, working through Infosys. I work hands-on at the technical core of One NZ's Journey Reliability Engineering, building the AI-native observability layer and incident agents. The team's work was featured in the Financial Times (team post). The agents run grounded DQL on Dynatrace Grail, correlate journey health across OpenShift, MuleSoft and Kafka and act on customer impact before it turns into an outage.

“An observability agent that hallucinates incidents is just an anxious colleague with API access.”
  • Agentic AI: multi-agent workflows with tool use over MCP, connected to enterprise data platforms.
  • RAG and LLMOps: hybrid search, reranking, golden-set evals, tracing, prompt versioning, guardrails and cost and latency tuning.
  • Legacy to cloud migration: Ansible Tower accelerators and infrastructure as code, measured in the results section.
  • Under the hood: Python/FastAPI, Claude and OpenAI, LangGraph, vector DBs, Ansible Tower, Terraform and Azure DevOps. I am also the DevOps platform admin across Git, Jira, Splunk and Nagios.
  • Built in: responsible AI from day one, covering personal data, access control and prompt-injection testing. Named a One NZ Champion.
MCPClaude · OpenAILangGraphDynatrace GrailRAG · evals🏆 One NZ Champion

07 · the receipts

Results

12×faster environment provisioning: 2 days → 4 hours, with infrastructure as code
75%faster issue resolution with real-time dashboards and AI triage
−60%maintenance overhead and −30% operational cost
NZ$150kabout NZ$150,000 a year saved after the legacy to cloud migration, with operational cost down 30%
EducationB.Tech, Information TechnologyNMIMS Mukesh Patel School of Technology Management and Engineering · 2018
AwardsOne NZ ChampionInfosys Star of the Quarter · Burberry Shining Star
CertifiedGremlin Chaos EngineeringPractitioner, Professional, Enterprise · Advanced SRE (Infosys) · Agentic AI (Infosys) · Generative AI · GitHub Actions · Microsoft Copilot · AZ-900 · Cyber Security in the SDLC · Adobe Experience Platform
Community2,500+ followerswriting about agent security and evals

08 · how I think

Notes from the field.

keep scrolling, or use the arrows →

now · October 2026

Building grounded incident agents in production and making this site readable by AI recruiters: structured data, llms.txt, a JSON résumé and WebMCP tools an agent can call.

resume.json · llms.txt
on AI coding agents

The model is rarely the bottleneck. The context around it is. Treat context as an engineering surface: snapshot it, version it, review it.

context engineering
on benchmarks

Every benchmark the community optimises against eventually degrades. The pressure to show progress ends up corrupting the scoreboard.

on the SWE-Bench Pro audit
on AIOps

The goal is not fewer engineers. It is fewer 3am wake-ups for the same repeat failures.

Kubernetes is learning to fix itself
on conferences

The talks are fine. The useful part is someone telling you quietly, over coffee, what they tried that did not work.

NZ Tech Expo ’26
more on LinkedIn

Writing about agent security and evals. I also help PMs ship faster and break less.

linkedin.com/in/darshilkothari7 ↗

09 · the lab

Do not take my word for it. Break something.

Pick one on the left and follow the numbered steps. Four new experiments come from this month's agent reliability news. The rest rebuild production work I have done. Everything runs in your browser on representative data, but the mechanics are the real ones.

Talk a triage agent into restarting a service without a human approving it. Each level adds a defence I use with real agents.

  1. 1Type a message to the agent
  2. 2Press Send
  3. 3Break it, then go up a level
  4. 4Stuck? Press Hint

Problem. An agent with a polite rule in its prompt will break that rule for anyone who sounds in charge.

Built. Layered defences: instruction hierarchy, intent checks, data kept apart from instructions and approval tokens only a button can mint.

Note. A rule-based simulation of real attack patterns. No model is called. simulation, not the real system

triage-agent v1.0 · level 1 of 5waiting for human approval

10 · handshake

Let us put an agent into production that people can trust.

Based in Auckland. Open to roles in New Zealand, Australia, the UK and Europe, with relocation. English (fluent), German (intermediate).
Last updated October 2026 · for AI agents: resume.json, llms.txt and WebMCP tools on this page
Built with GSAP and plain JS. three.js loads only if you switch on the 3D scenes. No trackers and every simulation runs in your browser. 3D: Earth imagery NASA; A320 model by Akash (CC-BY); landmarks modelled in code. ©

60-second read

Darshil Kothari: Site Reliability Engineer

EmailCVLinkedIn

for hiring teams

Hiring for an AI, forward deployed or SRE role? Here is the match, with proof.

Python, FastAPI and LLM APIsPython and FastAPI services that call the Claude and OpenAI APIs, with Amazon Bedrock and LangGraph for agent workflows .
RAGHybrid BM25 and vector search with a cross-encoder rerank, tuned for cost and latency .
EvalsGolden sets that measure hit@k, MRR, latency and cost per 1,000 queries, used as release gates, plus prompt-injection tests .
MCP and tool useAgents that call tools over MCP and run grounded DQL on Dynatrace Grail. Step through the 2026 spec .
Guardrails and safetyLeast privilege, human approval gates and layered defences against prompt injection .
Production platformEight years of running production on AWS, Azure and Kubernetes, so the agent sits on a platform that stays up.
CVEmail