Using LLMs to Solve Workflow Bottlenecks

I’ve been building and managing data systems at Amazon for the last 8 years. Now that AI is everywhere, the way we work as data engineers is changing fast. Here are 5 real ways I (and many in the industry) use LLMs to work smarter every day as a Senior Data Engineer: 1. Code Review and Refactoring LLMs help break down complex pull requests into simple summaries, making it easier to review changes across big codebases. They can also identify anti-patterns in PySpark, SQL, and Airflow code, helping you catch bugs or risky logic before it lands in prod. If you’re refactoring old code, LLMs can point out where your abstractions are weak or naming is inconsistent, so your codebase stays cleaner as it grows. 2. Debugging Data Pipelines When Spark jobs fail or SQL breaks in production, LLMs help translate ugly error logs into plain English. They can suggest troubleshooting steps or highlight what part of the pipeline to inspect next, helping you zero in on root causes faster. If you’re stuck on a recurring error, LLMs can propose code-level changes or optimizations you might have missed. 3. Documentation and Knowledge Sharing Turning notebooks, scripts, or undocumented DAGs into clear internal docs is much easier with LLMs. They can help structure your explanations, highlight the “why” behind key design choices, and make onboarding or handover notes quick to produce. Keeping platform wikis and technical documentation up to date becomes much less of a chore. 4. Data Modeling and Architecture Decisions When you’re designing schemas, deciding on partitioning, or picking between technologies (like Delta, Iceberg, or Hudi), LLMs can offer quick pros/cons, highlight trade-offs, and provide code samples. If you need to visualize a pipeline or architecture, LLMs can help you draft Mermaid or PlantUML diagrams for clearer communication with stakeholders. 5. Cross-Team Communication When collaborating with PMs, analytics, or infra teams, LLMs help you draft clear, focused updates, whether it’s a Slack message, an email, or a JIRA comment. They’re useful for summarizing complex issues, outlining next steps, or translating technical decisions into language that business partners understand. LLMs won’t replace data engineers, but they’re rapidly raising the bar for what you can deliver each week. Start by picking one recurring pain point in your workflow, then see how an LLM can speed it up. This is the new table stakes for staying sharp as a data engineer.

LLMs are the single fastest way to make yourself indispensable and give your team a 30‑percent productivity lift. Here is the playbook. Build a personal use‑case portfolio Write down every recurring task you handle for clients or leaders: competitive intelligence searches, slide creation, meeting notes, spreadsheet error checks, first‑draft emails. Rank each task by time cost and by the impact of getting it right. Start automating the items that score high on both. Use a five‑part prompt template Role, goal, context, constraints, output format. Example: “You are a procurement analyst. Goal: draft a one‑page cost‑takeout plan. Context: we spend 2.7 million dollars on cloud services across three vendors. Constraint: plain language, one paragraph max. Output: executive‑ready paragraph followed by a five‑row table.” Break big work into a chain of steps Ask first for an outline, then for section drafts, then for a fact‑check. Steering at each checkpoint slashes hallucinations and keeps the job on‑track. Blend the model with your existing tools Paste the draft into Excel and let the model write formulas, then pivot. Drop a JSON answer straight into Power BI. Send the polished paragraph into PowerPoint. The goal is a finished asset, not just a wall of text. Feed the model your secret sauce Provide redacted samples of winning proposals, your slide master, and your company style guide. The model starts producing work that matches your tone and formatting in minutes. Measure the gain and tell the story Track minutes saved per task, revision cycles avoided, and client feedback. Show your manager that a former one‑hour job now takes fifteen minutes and needs one rewrite instead of three. Data beats anecdotes. Teach the team Run a ten‑minute demo in your weekly stand‑up. Share your best prompts in a Teams channel. Encourage colleagues to post successes and blockers. When the whole team levels up, you become known as the catalyst, not the cost‑cutting target. If every person on your team gained back one full day each week, what breakthrough innovation would you finally have the bandwidth to launch? What cost savings could you achieve? What additional market share could you gain?

Building LLM Agent Architectures on AWS - The Future of Scalable AI Workflows What if you could design AI agents that not only think but also collaborate, route tasks, and refine results automatically? That’s exactly what AWS’s LLM Agent Architecture enables. By combining Amazon Bedrock, AWS Lambda, and external APIs, developers can build intelligent, distributed agent systems that mirror human-like reasoning and decision-making. These are not just chatbots - they’re autonomous, orchestrated systems that handle workflows across industries, from customer service to logistics. Here’s a breakdown of the core patterns powering modern LLM agents : Breakdown: Key Patterns for AI Workflows on AWS 1. Prompt Chaining / Saga Pattern Each step’s output becomes the next input — enabling multi-step reasoning and transactional workflows like order handling, payments, and shipping. Think of it as a conversational assembly line. 2. Routing / Dynamic Dispatch Pattern Uses an intent router to direct queries to the right tool, model, or API. Just like a call center routing customers to the right department — but automated. 3. Parallelization / Scatter-Gather Pattern Agents perform tasks in parallel Lambda functions, then aggregate responses for efficiency and faster decisions. Multiple agents think together — one answer, many minds. 4. Saga / Orchestration Pattern Central orchestrator agents manage multiple collaborators, synchronizing tasks across APIs, data sources, and LLMs. Perfect for managing complex, multi-agent projects like report generation or dynamic workflows. 5. Evaluator / Reflect-Refine Loop Pattern Introduces a feedback mechanism where one agent evaluates another’s output for accuracy and consistency. Essential for building trustworthy, self-improving AI systems. AWS enables modular, event-driven, and autonomous AI architectures, where each pattern represents a step toward self-reliant, production-grade intelligence. From prompt chaining to reflective feedback loops, these blueprints are reshaping how enterprises deploy scalable LLM agents. #AIAgents

I started by asking AI to do everything. Six months later, 65% of my agent’s workflow nodes run as non-AI code. The first version was fully agentic : every task went to an LLM. LLMs would confidently progress through tasks, though not always accurately. So I added tools to constrain what the LLM could call. Limited its ability to deviate. I added a Discovery tool to help the AI find those tools. Better, but not enough. Then I found Stripe’s minion architecture. Their insight : deterministic code handles the predictable ; LLMs tackle the ambiguous. I implemented blueprints, workflow charts written in code. Each blueprint specifies nodes, transitions between them, trigger conditions for matching tasks, & explicit error handling. This differs from skills or prompts. A skill tells the LLM what to do. A blueprint tells the system when to involve the LLM at all. Each blueprint is a directed graph of nodes. Nodes come in two types : deterministic (code) & agentic (LLM). Transitions between nodes can branch based on conditions. Deal pipeline updates, chat messages, & email routing account for 29% of workflows, all without a single LLM call. Company research, newsletter processing, & person research need the LLM for extraction & synthesis only. Another 36%. The workflow runs 67-91% as code. The LLM sees only what it needs : a chunk of text to summarize, a list to categorize, processed in one to three turns with constrained tools. Blog posts, document analysis, bug fixes are genuinely hybrid. 21% of workflows. Multiple LLM calls iterate toward quality. Only 14% remain fully agentic. Data transforms & error investigations. These tend to be coding tasks rather than evaluating a decision point in a workflow. The LLM needs freedom to explore. AI started doing everything. Now it handles routing, exceptions, research, planning, & coding. The rest runs without it. Is AI doing less? Yes. Is the system doing more? Also yes. The blueprints, the tools, the skills might be temporary scaffolding. With each new model release, capabilities expand. Tasks that required deterministic code six months ago might not tomorrow.

We know LLMs can substantially improve developer productivity. But the outcomes are not consistent. An extensive research review uncovers specific lessons on how best to use LLMs to amplify developer outcomes. 💡 Leverage LLMs for Improved Productivity. LLMs enable programmers to accomplish tasks faster, with studies reporting up to a 30% reduction in task completion times for routine coding activities. In one study, users completed 20% more tasks using LLM assistance compared to manual coding alone. However, these gains vary based on task complexity and user expertise; for complex tasks, time spent understanding LLM responses can offset productivity improvements. Tailored training can help users maximize these advantages. 🧠 Encourage Prompt Experimentation for Better Outputs. LLMs respond variably to phrasing and context, with studies showing that elaborated prompts led to 50% higher response accuracy compared to single-shot queries. For instance, users who refined prompts by breaking tasks into subtasks achieved superior outputs in 68% of cases. Organizations can build libraries of optimized prompts to standardize and enhance LLM usage across teams. 🔍 Balance LLM Use with Manual Effort. A hybrid approach—blending LLM responses with manual coding—was shown to improve solution quality in 75% of observed cases. For example, users often relied on LLMs to handle repetitive debugging tasks while manually reviewing complex algorithmic code. This strategy not only reduces cognitive load but also helps maintain the accuracy and reliability of final outputs. 📊 Tailor Metrics to Evaluate Human-AI Synergy. Metrics such as task completion rates, error counts, and code review times reveal the tangible impacts of LLMs. Studies found that LLM-assisted teams completed 25% more projects with 40% fewer errors compared to traditional methods. Pre- and post-test evaluations of users' learning showed a 30% improvement in conceptual understanding when LLMs were used effectively, highlighting the need for consistent performance benchmarking. 🚧 Mitigate Risks in LLM Use for Security. LLMs can inadvertently generate insecure code, with 20% of outputs in one study containing vulnerabilities like unchecked user inputs. However, when paired with automated code review tools, error rates dropped by 35%. To reduce risks, developers should combine LLMs with rigorous testing protocols and ensure their prompts explicitly address security considerations. 💡 Rethink Learning with LLMs. While LLMs improved learning outcomes in tasks requiring code comprehension by 32%, they sometimes hindered manual coding skill development, as seen in studies where post-LLM groups performed worse in syntax-based assessments. Educators can mitigate this by integrating LLMs into assignments that focus on problem-solving while requiring manual coding for foundational skills, ensuring balanced learning trajectories. Link to paper in comments.

Workflow v.s. AI Agents II: Get the Best of Both Worlds In my last post, I unpacked the differences between workflow systems and agentic systems and showed how both have propelled contact‑center AI forward. Each comes with clear pros, cons, and use‑case sweet spots. Today, I want to describe two patterns I’m seeing in real‑world deployments that capture the best of both worlds. 1. Workflow as a Tool to AI Agents Think of refund or authentication flows: you need them to be reliable, precise, and deterministic, no imagination, no exceptions. The right approach is to wrap each of those flows in code and let the LLM call it only when the conversation reaches the correct step. It’s the same strategy an LLM uses when it calls a calculator. The model handles natural language, then hands off to deterministic code. Because these calls rarely exist in isolation, you also maintain a lightweight global‑state store, e.g. customer ID, authentication status (e.g. failed codeword, 2nd attempt, need last 4 digits of SSN) , open‑case number, refund amount, and so on. Both the agent and the workflow read from and write to that state, so every turn starts on the same page. 2. Agentic System as a Fallback‑and‑Healing Layer Rule‑based workflows dominate high‑volume, repetitive back‑office tasks.  An invoice‑processing pipeline is a classic example, because cost and reliability matter more than creativity. The problem is that even the most battle‑hardened workflow eventually hits an edge case: an OCR misreads a field, a vendor changes a PDF layout, or a UI update moves a button or turns one text field into a drop down box. When that happens, route the exception to an LLM‑powered agent. The workflow raises a “can’t‑proceed” flag and passes the partial context. The agent reasons through the anomaly: asks a clarifying question, consults a knowledge base, rewrites the input, or tries to process the updated UI with an vLLM action model. The agent writes the corrected data back to the global state, then nudges the original workflow to resume. In effect, the deterministic layer handles the 95 % happy path, while the agentic layer patches the 5 % that rule‑based code can’t anticipate, and every successful patch becomes new training data for further hardening. In my next post, I will talk about test case management, evaluation to achieve determinism over underlying probabilistic models.

You don’t need to be an AI agent to be agentic. No, that’s not an inspirational poster. It’s my research takeaway for how companies should build AI into their business. Agents are the equivalent of a self-driving Ferrari that keeps driving itself into the wall. It looks and sounds cool, but there is a better use for your money. AI workflows offer a more predictable and reliable way to sound super cool while also yielding practical results. Anthropic defines both agents and workflows as agentic systems, specifically in this way: 𝗪𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀: systems where predefined code paths orchestrate the use of LLMs and tools 𝗔𝗴𝗲𝗻𝘁𝘀: systems where LLMs dynamically decide their own path and tool uses For any organization leaning into Agentic AI, don’t start with agents. You will just overcomplicate the solution. Instead, try these workflows from Anthropic’s guide to effectively building AI agents: 𝟭. 𝗣𝗿𝗼𝗺𝗽𝘁-𝗰𝗵𝗮𝗶𝗻𝗶𝗻𝗴:  The type A of workflows, this breaks a task down into sequential tasks organized and logical steps, with each step building on the last. It can include gates where you can verify the information before going through the entire process. 𝟮. 𝗣𝗮𝗿𝗮𝗹𝗹𝗲𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻: The multi-tasker workflow, this separates tasks across multiple LLMs and then combines the outputs. This is great for speed, but also collects multiple perspectives from different LLMs to increase confidence in the results. 𝟯. 𝗥𝗼𝘂𝘁𝗶𝗻𝗴: The task master of workflows, this breaks down complex tasks into different categories and assigns those to specialized LLMs that are best suited for the task. Just like you don’t want to give an advanced task to an intern or a basic task to a senior employee, this find the right LLM for the right job. 𝟰. 𝗢𝗿𝗰𝗵𝗲𝘀𝘁𝗿𝗮𝘁𝗼𝗿-𝘄𝗼𝗿𝗸𝗲𝗿���: The middle manager of the workflows, this has an LLM that breaks down the tasks and delegates them to other LLMs, then synthesizes their results. This is best suited for complex tasks where you don’t quite know what subtasks are going to be needed. 𝟱. 𝗘𝘃𝗮𝗹𝘂𝗮𝘁𝗼𝗿-𝗼𝗽𝘁𝗶𝗺𝗶𝘇𝗲𝗿: The peer review of workflows, this uses an LLM to generate a response while another LLM evaluates and provides feedback in a loop until it passes muster. View my full write-up here: https://lnkd.in/eZXdRrxz

🚀 Stop forcing one LLM to do everything, it’s time to hire a digital team. . . . . The industry often assumes a single, powerful model can handle complex reasoning and execution. In practice, however, one model trying to manage multiple data sources and distinct operations simultaneously often results in architectural failure. While a single agent may handle simple tasks instantly, it frequently breaks down when faced with complex, interconnected problems. ✅ Specialization Over Generalization: Distribute work across specialized agents (e.g., separate agents for billing, logistics, and recommendations) to maintain a focused context and reduce hallucinations. ✅ Validation via Peer Review: Multi-agent systems can self-correct through "orthogonal checking," where specialized agents cross-validate each other's outputs. ✅ Parallel Processing for Scale: Divide large data volumes among multiple workers to process them simultaneously, reducing a 20-minute task to just 3 minutes. ✅ Graceful Degradation: Unlike single-agent systems that suffer complete failure if one component crashes, multi-agent architectures can continue operating with partial results or spawn backup agents. ✅ Dynamic Cost Routing: Use lightweight, cheaper models for simple FAQs and reserve premium reasoning models for the 5% of queries that actually need them. The shift from a single "black box" model to a team of specialized agents isn't just about power it's about building a resilient, observable, and cost-effective digital workforce. Are you still trying to solve every complexity with better prompts, or have you started exploring multi-agent architectures? What's the biggest bottleneck you've faced with single-model systems? Source: Mastering Multi-Agent Systems (Galileo v1.01) 👉 Follow Sarveshwaran Rajagopal for more insights on AI, LLMs & GenAI. 🌐 Learn more at: https://lnkd.in/d77YzGJM #AI #LLM #MultiAgentSystems #GenAI #AgenticAI #MachineLearning #AIStrategy

Quick tip: you can build role-based agents in Claude to plan and execute work in parallel. Instead of one giant prompt, define lightweight agents for different functions (PM, dev, QA, UX, tech lead) and let an orchestrator coordinate them. When wired correctly, Claude can: - Plan work - Spin up sub-agents for specific tasks - Run things in parallel (not just sequential prompts) - Hand off context cleanly between roles - Support both sync and async workflows A very simple starting point: “Claude, build me agents (not commands) for .claude that simulate PM, dev, and QA roles. They should support sync and async work, include an orchestrator, and coordinate task execution.” Claude will rough these out surprisingly well. From there you can enhance: - Handoffs - Session management - Resume logic - Guardrails - Domain-specific behavior Think of it less like prompting… and more like assembling a small delivery team that knows how to talk to each other. This pattern has quietly changed how I approach planning and execution. If you’re still treating LLMs as single-shot tools, this is worth experimenting with Thanks to Nirav Sheth for reminding me about this powerhouse set of tools 🎉

🚀 3 ways I use LLMs to accelerate the process of taking any workstream step from 0 to 1 (idea to execution): 👼 By the way, I’m an investor in OpenAI (ChatGPT), Anthropic (Claude), and Perplexity. So I’m truly ‘invested’ in understanding how to make the most of these tools. 📈 1. Creating a Growth Plan I’ll upload several growth audits & growth plans from previous projects and ask the LLM to create a new template tailored to a different business model or GTM motion. Then I’ll edit and fill it in using my specific expertise. 📢 2. Creating a Comms Doc Much like a growth plan, I’ll upload several comms docs from my product marketing days as well as the product brief a client is delivering against. Then I’ll ask the LLM to generate a comms doc for this new product that I can edit as I see fit. 📃 3. Customer Feedback Summaries Customer interviews are hugely valuable, but time-consuming to summarise. I’ll upload the transcripts from 10 customer interviews along with a summary template and ask the LLM to count the number of times specific themes are mentioned. ___ I believe all knowledge workers should be using LLMs to stay relevant and scale themselves The key is to document all of your principles, knowledge, and playbooks. 🖥️ And feed it all into an LLM. (If you’re worried about privacy, NotebookLM and Anthropic say they don’t train on your personal data, and you can opt out of OpenAI model training.) ✅ You’ll get about 80% of the easy work done in two minutes. Things like: — Tone and style matching — Flagging anything you’re missing — Page formatting & content structuring  — Generating a template for something you’ve never done before 🧠 And you’re just left to do the thoughtful, valuable, expert-level work that helps accelerate growth.