AI Agents Developer (LLM/ML Engineer)

AI Agents Developer (Remote, Flexible) — LLM / ML Engineer

Position: AI Agents Developer (LLM / ML Engineer)

Work Type: Remote (Flexible hours)

Employment: Full-time/Part Time

Level: Junior-to-Mid (Strong fundamentals required; fast growth path)

Product Focus: AI Agents + AI SaaS Products & Services

We’re building AI Agents and AI SaaS product services that automate real business workflows end-to-end research, planning, tool usage, reporting, and integrations. Our focus is not “demo AI.” We ship production systems that are reliable, measurable, and scalable.

If you’re obsessed with LLMs, agent frameworks, automation, and building real systems then this role is for you.

Compensation & Rewards

•Starting Salary: BDT 60,000 – 70,000 / month (starter range, based on skills)

•1st Increment: After 6 months

•Yearly Bonus: Performance-based annual bonus

•Revenue Sharing Opportunity: Exceptional contributions can earn revenue sharing (rewarded based on impact and product growth)

•Relocation Path: After 2 years of successful performance, opportunity to move to the Netherlands (role dependent + business needs)

What You’ll Build

•Developing high performing AI agents that can plan, execute multi-step tasks, and call APIs reliably

•LLM pipelines: structured outputs, function calling, guardrails, retries, self-correction

•RAG systems: chunking, embeddings, reranking, hybrid search, citation grounding

•Agent architectures: planner/executor/critic patterns, routing, memory, task decomposition

•AI SaaS services: backend APIs, integrations, workflows, dashboards/analytics

•Evaluation & observability: agent test suites, regression checks, logging/tracing

Key Responsibilities

•Build and improve AI agents that complete real workflows with high accuracy

•Integrate LLMs with tools: web/search, internal APIs, databases, files, spreadsheets

•Create robust prompt + schema systems (JSON validation, tool-call correctness)

•Develop RAG pipelines and improve grounding to reduce hallucinations

•Set up and maintain evaluation: benchmarks, test cases, automated checks

•Optimize latency, cost, and reliability for production use

•Work closely with product to ship features quickly and cleanly

Must-Have Skills (We will test)

LLM + Agent Development

•Strong knowledge of LLMs and how to build reliable agent loops

•Tool/function calling, routing, structured JSON outputs, guardrails

•Understanding of hallucination risks + grounding strategies

AI/ML Foundations

•Solid ML/NLP fundamentals (embeddings, similarity, retrieval, evaluation metrics)

•Ability to design experiments and measure improvements

Production Engineering

•Strong Python (required)

•APIs (FastAPI/Flask), async jobs/queues, clean code practices

•Git, debugging, performance tuning

Retrieval & Data (Preferred if you have)

•Vector DBs (FAISS / Qdrant / Pinecone / Weaviate / pgvector)

•SQL (Postgres), caching (Redis), background workers (Celery/RQ)

Nice-to-Have (Big Advantage)

•Experience with LangChain / LlamaIndex / similar agent frameworks

•Experience deploying AI systems (Docker, cloud, CI/CD)

•Knowledge of security basics: prompt injection defense, PII handling, safe tool access

•TypeScript/Node (helpful for integrations and tooling)

Who Should Apply

You should apply if you:

•Can build real systems (not only prompts)

•Love shipping fast and improving with feedback

•Want growth + potential revenue upside + future relocation opportunity

•Are comfortable owning features end-to-end

Work Style

•Remote + flexible hours

•Outcome-driven: you manage your time, we care about results

•Fast iteration, strong engineering standards, real production shipping

Hiring Process (Simple & Practical)

1.Short technical interview (LLM + Python + systems)

2.Small task: build a mini agent with tool calling + structured output

3.Final discussion + offer

How to Apply

Send:

•CV / LinkedIn

•GitHub / portfolio (if available)

•Short message: what AI agent or LLM project you built (even personal projects)