METAHEURISTICBook a call
[Core role]

Hire a freelance agentic AI engineer.

Multi-step agents that plan, call tools, and stop when they should. Built with budgets, evals and traces from the first commit - not bolted on after the demo goes sideways.

2.0k
Stars on agentic_security
938
Stars on langcorn (LLMOps)

An agent is not a prompt. It is a control loop with money attached: a model decides, a tool runs, state changes, and the loop goes around again. Everything that makes agents hard is in that sentence - non-determinism at the decision point, side effects at the tool boundary, and unbounded cost in the loop.

Hiring an agentic AI engineer means hiring someone who treats that loop as a distributed system rather than a clever prompt.

What the work actually is

Control flow

Explicit state, explicit transitions, explicit termination. A step budget, a token budget and a wall-clock budget, all enforced by the harness rather than requested politely in the system prompt. No-progress detection so the agent notices when it has called the same tool with the same arguments three times.

Tool contracts

Tools are an API you are exposing to an unreliable caller. That means narrow signatures, validated arguments, idempotency keys on anything with a side effect, least-privilege scopes, and errors that are readable by a model rather than a stack trace. Most “the agent is dumb” bugs are tool-design bugs.

Memory and context

What the agent carries forward, what it re-derives, what it writes down. Context windows are a budget, not a bucket - and a blackboard the agent writes to beats stuffing every prior turn back into the prompt.

Evaluation

A golden set of tasks with pass criteria, run in CI, with a number attached to every change. Without it, prompt engineering is superstition and every release is a coin flip. With it, you can let a junior engineer touch the prompt.

Guardrails and red-teaming

Prompt injection through retrieved documents, tool abuse, exfiltration through tool arguments, jailbreaks that unlock a dangerous action. These get adversarial test cases in the same suite as the happy path.

AgentOps

Traces you can replay, cost per completed task rather than cost per token, a failure taxonomy that tells you which 5% of runs are eating 60% of the spend, and a rollback that takes a deploy rather than an incident.

Typical engagements

ShapeLengthWhat comes out
Agent architecture review1-2 weeksWritten teardown, failure taxonomy, prioritised fix list, cost model
Discovery Sprint2-4 weeksWorking thin slice, architecture, eval harness, security pre-check, roadmap
Embedded build4-10 weeksProduction agent behind flags, evals in CI, traces, runbooks, handover
FractionalRecurringSenior AI judgement in your architecture reviews, hiring loops and roadmap

Stack I reach for

Python and Go. Claude, GPT and open-weight models behind a routing layer so no single provider is load-bearing. LangGraph or hand-rolled state machines. Temporal when the workflow must survive a restart. Postgres and pgvector before anything more exotic. OpenTelemetry for traces. Docker and Kubernetes, on your cloud or on-premise.

The stack is negotiable. The evals, budgets and traces are not.

Why me specifically

I maintain agentic_security, an LLM vulnerability scanner and red-teaming kit with around 2,000 GitHub stars, and langcorn, which serves LangChain apps and agents through FastAPI. The failure modes on this page are not read from a vendor blog - they came from building the tools other teams use to find them.

I also write about this work in public, roughly weekly, with the argument and the counter-argument both on the page.

Questions before you hire.

Which agent framework do you use? +

Whichever survives contact with your constraints - and often none. LangGraph, the Anthropic and OpenAI SDKs, Temporal for durable execution, or plain Python with a state machine. Frameworks are a convenience, not an architecture; the control flow, tool contracts and eval harness are the parts that decide whether the thing works.

We already have an agent that half works. Can you fix it? +

That is the common case. The usual diagnosis is no termination budget, tools with ambiguous contracts, no eval set, and no trace you can replay. Those are fixable in weeks, not quarters, and the fix is mostly deletion.

How do you stop an agent from looping or burning budget? +

Step ceilings, wall-clock and token budgets enforced at the harness rather than in the prompt, no-progress detection on repeated tool calls, idempotent tools, and a hard cutoff that escalates to a human instead of retrying forever. I have written about tool-call cycles at length.

What about prompt injection and tool abuse? +

Treated as a threat model: least-privilege tool scopes, allowlisted side effects, output validation before anything writes, and adversarial test cases in the eval suite. I maintain agentic_security, an open-source red-teaming kit for exactly this class of bug.

Do you work with our existing engineers? +

Preferably. The best outcome is your team owning the system at the end. I pair, review, and leave the harness, evals and runbooks behind.

Need an agentic AI engineer this quarter?

Contract, fractional, or embedded with your team. Start with a 20-minute call - you leave it with an honest read on whether your use case is buildable, whatever happens next.