Dobs.ai

Experience Level: 3-5+ years of overall engineering experience, including at least 1 year working with LLM-based agentic systems

About Us

We’re building a next-generation platform that leverages Large Language Models (LLMs) and agentic workflows to transform document processing, analytics, and decision-making for large interprises (Top Fortune 1000). Our fast-growing team is looking for an AI Engineer who will design and implement robust agent flows, integrate advanced LLM frameworks and design patterns, and optimize context-building from multiple data sources.

What You’ll Do

• Design & Implement Agentic Workflows
  • Build AI “agents” with memory modules, context-building, and tool integration.
  • Develop strategies for connecting diverse data sources (e.g., DB, Vector Stores, other MCPs) into agent pipelines.
• LLM Integration & Prompt Engineering
  • Craft effective prompts for chain-of-thoughts for domain-specific tasks (extraction, reasoning, summarization, data augmentation) to reduce hallucinations.
  • Evaluate and fine-tune OpenSource LLM like LLaMA models for specific business needs.
• Architecture & Best Practices
  • Drive architecture design patterns for robust, scalable AI microservices.
  • Apply best practices in software engineering (CI/CD, testing, code reviews).
• Data Pipelines & Context Handling
  • Build/maintain data workflows for ingesting and storing large volumes of unstructured/structured data.
  • Collaborate with data team to ensure high-quality data is fed into LLM pipelines.
• Collaborate Across Teams
  • Work with clients, analyst, product, backend, and devops teams to bring agentic solutions into production.

  • Share your expertise in design and optimization of AI-driven applications.

     

Must-Have Experience

✅ 3+ years of software engineering experience (Python preferred).

✅ 1+ year of direct experience building LLM-based agentic systems (memory, context-building, tool integration).

✅ Proven track record of prompt engineering and reducing hallucinations in LLM tasks.

✅ Experience with advanced retrieval-augmented generation (RAG) methods like vector databases, knowledge graphs, etc.

✅ Experience building scalable backend systems from scratch / 0-1 (not just maintenance).

✅ Strong understanding of design patterns (both traditional architecture and agentic flow).

✅ Hands-on approach - must verify work through unit testing & end-to-end evals.

Nice to Have

• Background in machine learning or data science (training and deploying models at scale).
• Understanding of domain-specific best practices (security, compliance, large-scale production environments).
• Familiarity with data pipelines (ingestion, transformation, storage) for unstructured/structured data.
• Comfortable with version control, CI/CD, testing frameworks, and containerization (e.g., Docker).

Key Responsibilities

1. Agent Development
   • Implement LLM-based agents with memory/context features and external tool integrations (e.g., DB queries, retrieval APIs).
2. Prompt Engineering
   • Experiment with chain-of-thought prompts, advanced reasoning flows, and consistent context injection.
3. Systems Architecture
   • Champion proven architecture design patterns for reliability, performance, and scalability.
   • Contribute to the evolution of the overall software stack (Python backend).
4. Continuous Improvement
   • Drive the iteration of agentic workflows, measure performance, and optimize real-time data pipelines.
5. LLM & ML Experimentation
   • (Nice to have) Use ML frameworks to develop ensemble or fallback solutions if LLM context is insufficient.

   • Evaluate model performance on internal validation datasets.

      

Why Join Us

• Impact & Ownership: Be a founding AI engineer spearheading the design of agentic workflows that push the boundaries of LLM usage.
• Cutting-Edge Tech: Work with the latest LLMs, advanced memory & context frameworks, and best-in-class software stacks.
• Growth Opportunities: Shape a core AI function in a fast-paced environment; collaborate with top talent across product, engineering, and data.
• Flexible Culture: We offer remote/hybrid setups with an emphasis on results over hours.

Ready to Build the Future of Agentic AI?

🚀 We look forward to hearing from you!