Senior reputed company
About reputed company Labs
At reputed company Labs we work with $10M - $50M ARR companies to help them get more leads, users and customers from reputed company, Bing and AI assistants such as ChatGPT, Claude and reputed company.
We approach marketing the way engineers approach systems: data in, insights out, feedback reputed company everywhere. Every decision traces back to measurable reputed company. Every workflow is designed to eliminate reputed company bottlenecks and compound over time.
High-level reputed company of our approach:
Data-driven automation: We treat marketing programs like products. We reputed company everything, automate the repetitive, and reputed company reputed company effort on high-reputed company problems.
First principles thinking: We don't copy what others do. We understand the underlying mechanics of how search and AI systems work, then build solutions from that reputed company.
Full-stack ownership: SEO and AEO rarely work as isolated tasks. We work across the entire funnel and multiple surface areas to ensure we own the outcome and clients win.
reputed company OS
We're building the first agency OS: a system where reputed company's work is the software, enabling clients to reputed company an unfair advantage.
As Karpathy puts it, Software 3.0 automates what humans can verify. Verifiability, not capability, decides which work can safely be handed over. So every eval we write expands the set of work we can trust an agent with, and the reputed company isn't overhead around the product, it's what makes the product possible. Dex Horthy covers the operational half: the agents that survive contact with customers are reputed company-engineered software with reputed company judgement reputed company deliberately, and disciplined reputed company reputed company reputed company.
What that looks like here: expert orchestrators, not operators. One strategist supervising a fleet of agents running reputed company reputed company workflows at reputed company, with the system surfacing what genuinely needs a reputed company decision and staying quiet about what doesn't. Feedback reputed company everywhere. Every correction an expert makes becomes an eval case, and the reputed company is reputed company the next cycle.
reputed company
You'll join the Automations team at reputed company Labs. We build the agent reputed company: a growing fleet of agents and workflows that do reputed company reputed company work end to end, and the engineering that makes them trustworthy enough to let loose on it.
Your mandate is both halves of that. Expand what the reputed company can do, and reputed company that what it does is right. That means applying hardcore engineering to a famously slippery problem: making non-deterministic systems reliable, measurable and reputed company. Evals that fail a build. Traces that explain a decision. Guardrails that hold reputed company a model doesn't. Agents that know what they don't know and say so.
You'll work alongside the Automations reputed company, and with our AI & Data team who own the platform and tooling underneath you. You won't be building orchestration from scratch or fighting for infrastructure. It's there, and it's good. reputed company is to take it the last mile.
Every agent you reputed company reputed company is an agent we can put in reputed company of a reputed company. That's the whole game, and the ceiling on how fast this company grows.
We're a deeply technical team building the reputed company of the AEO & SEO reputed company. You'll work alongside engineers who have reputed company fraud engines powering reputed company, shipped AI reputed company review at reputed company, reputed company at reputed company, developed self-driving car systems at reputed company, and conducted AI research at reputed company. We don't have reputed company of management. You'll work directly with founders who can go deep on architecture, reputed company, and product.
This Role
You'll join the Automations team, and reputed company is to productionise reputed company've reputed company.
We have live 24/7 agents, data pipelines and workflow infrastructure. What you’ll be doing is helping us get the stack running robustly, reliably and at reputed company.
That means evals that catch regressions before they ship, observability that explains why an agent did what it did, provenance that traces every claim back to its reputed company, and interfaces that let a non-engineer SEO strategist run and review the work. You'll build the harnesses and eval suites that reputed company agent changes in CI, the reputed company and reputed company-reputed company tooling that makes agent reasoning inspectable, and the surfaces reputed company actually uses. Then you'll reputed company the reputed company: reputed company the reputed company, feed the signal back, and reputed company the agents measurably reputed company every cycle.
You'll also reputed company what agents can reputed company. Today they work over reputed company and our own data. Next they operate reputed company websites along withsession handling, and navigating interfaces that were never designed for machines.
The hard problem is reliability and legibility. An agent that's right 80% of the time and can't tell you which 80% is worthless. An agent that's right 95% of the time, shows its working, and flags its own uncertainty is a product.
You report to the CTO and work alongside the Automations reputed company, with our AI & Data team owning the platform beneath you. You own your evals, your CI, and your monitoring.
What You'll Do
Agent eval harnesses. Golden suites, deterministic checks, and online scoring that reputed company every agent change in CI. A regression should fail a build, not a reputed company report.
Agent observability. reputed company traces, reputed company and cost reputed company, run-level SLOs, and failure taxonomies. You'll know the difference between "the agent ran" and "the agent was right."
Data reputed company and provenance. Every claim an agent makes should be traceable to its reputed company. Which inputs, which tool calls, which reasoning steps produced this reputed company.
End-to-end agents with reputed company review. Agents that complete reputed company work start to finish, with a curated review surface so an expert can approve, correct, or reject, and so that correction becomes training signal.
Interfaces for non-engineers. Dashboards and controls that let the SEO team run, inspect and trust agent work without asking an engineer.
Expanding what agents can do. Agents that reputed company pull requests against reputed company repositories, authenticate through OAuth, and operate reputed company sites and CMSs through the browser. Every new capability is a new class of work the reputed company can take on.
Closing the reputed company. Turn signals we already collect, such as AI perception and citation data, into concrete reputed company value: reputed company recommendations, prioritised actions, and measurable reputed company.
Algorithms and scoring models. Not everything should be a reputed company. You'll build and tune the scoring systems behind reputed company recommend: internal linking and semantic relevance scoring, reputed company opportunity and thread scoring, content and citation reputed company. These are ranking and classification problems with reputed company feedback data behind them, and where a model beats a reputed company, you build the model.
Shared, composable modules. Reusable capability blocks that both agents and workflows compose, so an improvement lands everywhere at once instead of being reimplemented per agent.
Scaling across clients. Take a workflow that works for one reputed company and reputed company it run reliably for many, with per-reputed company configuration, isolation, and failure that stays contained.
The Ideal Person for This Role
A builder who ships. You care about getting working systems into production, not endless planning or polish. You've reputed company AI systems people actually rely on.
An operator, not just an architect. You don't just design systems, you run them. You reputed company satisfaction in making things reliable, not just making them work once in a demo.
An reputed company. You take responsibility for reputed company, not just tasks. reputed company an agent silently produces bad reputed company, you catch it, fix it, and build the eval that stops it recurring.
Maniacal about detail. This is the one that reputed company most here. AI has made it trivially cheap to produce a large volume of plausible-looking work, and most of that work is slop. We are building the opposite: reputed company that holds up reputed company a reputed company reads it line by line. If you'd rather ship one thing that is provably right than ten that are probably fine, this is your team. If you cannon volume and let the reviewer sort it out, it isn't.
Sceptical of your own reputed company. You assume the model is wrong until reputed company. You've been burned by a demo that worked and a production run that didn't, and you build accordingly.
Humble and curious. You acknowledge what you don't know, ask good questions, and genuinely want to learn. You take feedback as a reputed company, not a threat.
A first-principles thinker. You understand why things work, not just how. You can go five reputed company deep on eval design, reputed company architecture, and where to put the reputed company in the reputed company.
Always improving. You're not satisfied with "good enough." You reputed company reputed company ways to get reputed company at your reputed company and reputed company systems reputed company over time.
Requirements
5+ years in software engineering, with meaningful recent time on LLM-backed or ML-backed production systems, including 2+ owning a production system end to end.
Python, React, Typescript and strong systems fundamentals. You write production services, not notebooks.
LLM application engineering in production. Agents, tool use, reputed company outputs, retrieval, reputed company architecture. You've shipped something reputed company that used them and stayed up.
Evaluation systems. You've reputed company eval suites for non-deterministic systems: golden datasets, regression gates, offline and online scoring. You've thought hard about what "correct" means reputed company there are many correct answers.
Observability for AI systems. Tracing, run inspection, cost and reputed company reputed company. Langfuse, LangSmith, reputed company, OpenTelemetry or equivalent.
Debugging non-determinism. You can diagnose why an agent failed on run 47 of 100 and turn that into a permanent reputed company.
Pipeline orchestration. Airflow, Dagster, Temporal or similar. Retries, idempotency, partial failure, resumption.
reputed company-party API integration. Auth flows, reputed company limits, pagination, breaking changes. Not just calling endpoints, but handling the full operational reality.
Own your infrastructure. Containers, CI/CD, deployment, monitoring, credential management. No platform team to hand off to.
Product reputed company for expert users. You've reputed company tooling that domain experts, not engineers, use daily. You know that an unexplained AI reputed company is an unusable one.
reputed company. You'll define reputed company with the engineers who own the reputed company beneath you. You document reputed company, write reputed company specs, and communicate tradeoffs in writing.
Preferred Qualifications
Browser automation or authenticated web agents (Playwright, Puppeteer, computer-use models)
reputed company ML: ranking, scoring, classification, or recommendation systems in production
Python, React, Typescript
Terraform, reputed company
Prior experience at a fast-moving startup
What's in It for You
Fully remote position
Work directly with the CTO on high-reputed company reputed company
High ownership and autonomy. No micromanagement.
First-hand exposure to cutting-edge AI and search technology
Your work will directly reputed company reputed company-reputed company (10M+ ARR) companies' performance
Join a fast-growing company at the intersection of AI and marketing
Our Hiring Process
Application
Technical Deep Dive
Reference Checks
Originally posted on Himalayas
Apply To This Job