Machine Learning Engineer, Agentic AI Solutions (Scientific Research Platform)

At Constructor Tech, we aim to revolutionize scientific discovery by empowering researchers with intelligent, agentic AI assistants. Our platform unifies literature mining, knowledge mapping, hypothesis generation, computational experimentation, results analysis, and publication support into a seamless, extensible environment. We’re seeking a seasoned ML engineer to help architect and build this next-generation research companion.

Role Overview

As a Machine Learning Engineer on our Agentic AI team, you will design, develop, and deploy the core platform that orchestrates autonomous AI agents and toolchains for diverse scientific workflows. You’ll collaborate closely with research scientists, data engineers, UX designers, and DevOps to turn cutting‑edge AI research into production‑grade features that accelerate literature review, knowledge graph construction, gap detection, computational modelling, and publication drafting.

Key Responsibilities:

• Platform Architecture & Development
• Architect and implement a modular, microservices‑based agentic AI platform supporting multi‑agent orchestration.
• Develop robust APIs and SDKs enabling seamless integration of AI assistants and external tools (e.g., literature databases, simulation engines).
• AI Agent & Tool Integration
• Build and integrate autonomous agents leveraging large language models (LLMs), retrieval‑augmented generation, and reinforcement learning for task planning and execution.

Incorporate specialized tools for:

• Literature Research: automated document retrieval, semantic search, summarization.
• Knowledge Mapping: dynamic knowledge graph construction, entity linking, relationship inference.
• Gap Finding & Hypothesis Generation: algorithmic identification of under‑explored research areas.
• Computational Research Pipelines: integration with simulation, statistical, and data‑analysis tools (e.g., Jupyter, SciPy, custom workflows).
• Results Analysis & Publication: data visualization modules, automated report and manuscript drafting.

• Model Development & Optimization
• Fine‑tune and benchmark LLMs, graph neural networks, and other deep learning architectures for domain‑specific tasks.
• Implement efficient inference pipelines, caching strategies, and batching for real‑time interactivity.
• Collaboration & Best Practices
• Work in cross‑functional Agile teams; participate in design reviews, sprint planning, and code reviews.
• Ensure high code quality, unit/integration testing, and continuous integration/deployment (CI/CD).
• Document system designs, APIs, and operational runbooks.

Required Qualifications:

• Education: Bachelor’s or Master’s in Computer Science, Machine Learning, AI, physics, chemistry or biology (PhD preferred).

Experience:

• 3+ years developing production‑scale ML/AI systems, ideally involving agentic or multi‑agent frameworks.
• Proven track record with LLMs (e.g., GPT, T5), RAG architectures, and knowledge graph technologies.

Technical Skills:

• Programming: Expert in Python; familiarity with Rust, Go, or TypeScript a plus.
• Frameworks & Libraries: PyTorch or TensorFlow; LangChain, LlamaIndex, Haystack, or similar.
• Data & Infrastructure: Elasticsearch, Neo4j or other graph databases; Docker, Kubernetes; AWS/GCP/Azure.
• Tooling: RESTful APIs, gRPC, message queues (e.g., RabbitMQ, Kafka).

Soft Skills:

• Strong problem‑solving, communication, and collaboration abilities.
• Comfort working in fast‑paced, research‑driven environments with evolving requirements.

Preferred Qualifications:

• PhD in AI/ML, Computational Science, or a scientific domain (biology, chemistry, physics).
• Experience in academic publishing or research support tools.
• Contributions to open‑source AI frameworks or scientific software.
• Familiarity with continuous learning pipelines and MLOps best practices.