#MLOps

Articles about MLOps — exploring patterns, best practices, and real-world implementations in production systems.

31 posts tagged with mlops. ← All posts

#A2A (14)#ADK (8)#AG-UI (6)#AI (9)#AI Agents (311)#AI Architecture (18)#AI Cost (10)#AI Cost Optimization (8)#AI Engineering (202)#AI Evaluation (8)#AI Governance (29)#AI Red Teaming (9)#AI Research (9)#AI Safety (8)#AI Security (29)#AI in Production (12)#AML (3)#API Design (10)#API Security (8)#APIs (55)#AWS (17)#Accounting (9)#Agent Skills (3)#Agentic AI (24)#Agentic Commerce (12)#Agentic RAG (8)#Agents (4)#Amazon Bedrock (8)#Analytics (3)#Architecture (37)#Audit (3)#Authentication (11)#Authorization (3)#Automation (8)#Azure (11)#Azure AI Foundry (9)#Backend Engineering (310)#Benchmarks (3)#Best Practices (3)#BigQuery (6)#Business Finance (8)#Business Strategy (55)#C (8)#CI/CD (16)#Caching (10)#Capital Markets (14)#Card Payments (12)#Cards (13)#Career (24)#Checkpointing (4)#Claude Code (8)#Cloud (5)#Cloud Architecture (3)#Cloud Native (10)#Code Review (8)#Collaboration (5)#Communication (9)#Compliance (52)#Computer Networking (9)#Computer Science (32)#Computer Vision (5)#Concurrency (39)#Consulting (3)#Containers (10)#Context Engineering (10)#Cost Optimisation (5)#Credit (14)#Credit Risk (14)#CrewAI (8)#Crypto (12)#Cryptocurrency (12)#Cryptography (8)#Custody (9)#DSPy (8)#Data (13)#Data Engineering (12)#Data Structures (9)#Databases (38)#Deployment (4)#Design Patterns (10)#DevOps (24)#DevSecOps (20)#Developer Experience (5)#Developer Tools (5)#Distributed Systems (95)#Documentation (3)#Edge AI (8)#Embeddings (15)#Emotional Intelligence (8)#Energy (8)#Engineering (11)#Engineering Culture (3)#Engineering Practices (16)#Error Handling (4)#Evaluation (54)#Event-Driven Architecture (8)#FREE-AI (8)#FX (5)#Feedback (4)#FinOps (23)#FinTech (6)#Financial AI (14)#Financial Systems (129)#Fine-Tuning (11)#Fintech (131)#Flutter (8)#Foreign Exchange (5)#Forward Deployed Engineer (8)#Forward Deployed Engineering (8)#Fraud (10)#Function Tools (5)#Functional Programming (3)#Fundraising (8)#GCP (5)#Gemma (4)#Generative AI (3)#Git (8)#Go (220)#Go-to-Market (8)#Google ADK (36)#Governance (57)#Granite (6)#GraphQL (3)#Growth (3)#Guardrails (33)#HIPAA (3)#HTTP (3)#Harness Engineering (8)#Hiring (8)#Hugging Face (8)#Human-in-the-Loop (8)#IBM watsonx (8)#Identity (11)#Integration (3)#Intellectual Property (8)#Interfaces (3)#JavaScript (8)#KYC (11)#KYC and AML (12)#Kafka (10)#Kubernetes (17)#LLM (5)#LLM Inference (8)#LLM-as-Judge (3)#LLMs (151)#LangChain (8)#LangGraph (11)#Leadership (26)#Ledger (12)#Legal (8)#Lending (14)#LlamaIndex (8)#MCP (22)#MLOps (31)#Machine Learning (40)#Marketing (16)#Markets (4)#Memory (15)#Memory Management (5)#Metrics (6)#Microservices (3)#Microsoft Agent Framework (150)#Middleware (6)#Migration (9)#Mixture of Experts (5)#Monitoring (3)#Multi-Agent (10)#Multi-Agent AI (14)#Multi-Agent Systems (73)#Multimodal (3)#Multimodal AI (8)#NIM (5)#NVIDIA (8)#OAuth (3)#OWASP (7)#Observability (45)#On-Device AI (8)#Open Source (7)#OpenTelemetry (5)#Operating Systems (8)#Operations (10)#Opinion (6)#Orchestration (10)#Organizational Design (8)#Payment Rails (16)#Payments (54)#People (8)#Performance (44)#Platform Engineering (9)#PreSales (8)#Privacy (5)#Privacy Engineering (3)#Process (4)#Product (27)#Product Management (8)#Production (8)#Programming (10)#Programming Languages (48)#Prompt Engineering (74)#Prompt Injection (14)#Protocol Buffers (3)#Protocols (9)#Providers (4)#Pydantic AI (8)#Python (142)#Quality (3)#RAG (58)#RBI (3)#REST (5)#Rails (16)#Reasoning Models (8)#Reconciliation (3)#RegTech (8)#Regulation (9)#Reliability (48)#Resilience (4)#Responsible AI (5)#Retrieval (3)#Risk (13)#Rust (32)#SLSA (3)#SRE (22)#Sales (9)#Scalability (3)#Security (89)#Security Engineering (8)#Self-Evolving Agents (16)#Sessions (3)#Settlement (9)#Soft Skills (8)#Software (3)#Software Architecture (36)#Software Delivery (9)#Software Engineering (144)#Spanner (4)#Startups (30)#Strands (8)#Streaming (28)#Structured Output (4)#Supply Chain Security (9)#Sustainability (8)#System Design (32)#Systems Programming (48)#Testing (52)#Threat Modeling (3)#Tool Use (22)#Tooling (4)#Tools (3)#Trading (8)#Treasury (6)#Type System (3)#Type Systems (11)#TypeScript (8)#Vector Databases (21)#Vector Search (11)#Venture Capital (8)#Version Control (8)#Web Development (6)#Workflows (14)#gRPC (13)#smolagents (8)
Pratik Dhanave · ·7 min read

AI Developer Tooling and the Provider Landscape

The AI bill isn't only your production inference — it's also the AI coding tools your engineers use all day, the gateways that route your traffic, and the accelerator and data-platform costs underneath. This closing post covers the spend beyond core inference and pulls the whole playbook into what generalizes across every provider.

The AI bill isn't only production inference — it's the coding tools your engineers use all day, the gateways that route your traffic, and the accelerator and data-platform costs underneath. This closes the playbook with what generalizes across every provider.

Pratik Dhanave · ·6 min read

LangChain in Production

The gap between a LangChain demo and a LangChain production system is the same gap as any LLM application — observability, evaluation, cost, and reliability — and LangChain's answer is LangSmith plus the discipline the rest of your engineering already has. This closing post covers operating LangChain applications and the honest verdict on when to use the framework.

The gap between a LangChain demo and a production system is the same as any LLM application — observability, evaluation, cost, and reliability — and LangChain's answer is LangSmith plus the discipline your engineering already has.

Pratik Dhanave · ·6 min read

Pydantic AI in Production

A framework earns its place not in the demo but in production — under real traffic, real failures, and the need to observe, control cost, and evolve. Pydantic AI's typed, testable design carries into production well, and paired with observability and the model-agnosticism it's had all along, it makes agents you can actually operate. This closing post covers taking a Pydantic AI agent live.

A framework earns its place in production — under real traffic, failures, and the need to observe, control cost, and evolve. Pydantic AI's typed, testable design carries into production well, paired with observability and model-agnosticism.

Pratik Dhanave · ·6 min read

Observability in Practice

Knowing the pillars is not the same as having an observable system. In practice, observability is built incrementally, costs real money you have to manage, and only pays off if the whole team treats telemetry as part of building software — not something added after the outage. This closing post turns the concepts into a way of working.

Knowing the pillars is not the same as having an observable system. In practice, observability is built incrementally, costs real money you must manage, and only pays off if the whole team treats telemetry as part of building software — not something added after the outage.

Pratik Dhanave · ·6 min read

Fine-Tuning in Production

Training a good fine-tune is the easy half. Running it in production — deciding it's even worth it, serving adapters efficiently, keeping it from going stale as base models leap ahead, and re-tuning as your needs shift — is where fine-tuning becomes an ongoing commitment rather than a one-time project. This is the reality check that closes the series.

Training a good fine-tune is the easy half. Running it in production — deciding it's worth it, serving adapters efficiently, keeping it from going stale as base models leap ahead, and re-tuning — is where fine-tuning becomes an ongoing commitment.

Pratik Dhanave · ·6 min read

Latency, Throughput, and Cost

There is no single "make it fast" for LLM serving — latency and throughput pull against each other, and both trade against cost. The job isn't to maximize one number; it's to hit your latency targets at the lowest cost per token, which means knowing exactly which knob moves which metric and in which direction.

There is no single 'make it fast' for LLM serving — latency and throughput pull against each other, and both trade against cost. The job is to hit your latency targets at the lowest cost per token, knowing which knob moves which metric.

Pratik Dhanave · ·4 min read

LlamaIndex in Production

A four-line RAG demo and a production RAG system share almost no operational concerns. Getting to production means treating ingestion as a pipeline, retrieval quality as something you measure, and the whole system as something you observe — the work that starts after the demo impresses everyone.

A four-line RAG demo and a production RAG system share almost no operational concerns. Getting to production means treating ingestion as a pipeline, retrieval quality as something you measure, and the whole system as something you observe.

Pratik Dhanave · ·5 min read

CrewAI in Production

CrewAI makes it easy to build a multi-agent system and just as easy to build one that's slow, expensive, and unreliable — so production CrewAI is mostly about discipline: measure it, keep it as simple as the problem allows, and use Flows for the parts that must be dependable.

CrewAI makes it easy to build a multi-agent system and just as easy to build one that's slow, expensive, and unreliable — so production CrewAI is mostly discipline: measure it, keep it simple, and use Flows where it must be dependable.

Pratik Dhanave · ·6 min read

Evaluation in Production

Offline evaluation tells you whether a change is promising; production tells you whether it actually works. Once your system is serving real users, evaluation becomes continuous: online experiments, guardrail metrics, drift monitoring, and gating deploys on eval scores. This closing post moves evaluation from the lab into the running system and ties the whole series into a working loop.

Offline evaluation tells you whether a change is promising; production tells you whether it works. Once you're serving real users, evaluation becomes continuous — online experiments, guardrail metrics, drift monitoring, CI gating. This closing post moves evaluation from the lab into the running system and ties the series into one loop.

Pratik Dhanave · ·6 min read

Evaluating a Fine-Tuned Model

A fine-tune that looks great on a handful of hand-picked prompts can be quietly broken — overfit to your training data, worse than the base model you started from, or catastrophically forgetful of skills it used to have. Without real evaluation you can't tell, and shipping a fine-tune you haven't measured is shipping a guess.

A fine-tune that looks great on a handful of prompts can be quietly broken — overfit, worse than the base model, or catastrophically forgetful. Without real evaluation you can't tell, and shipping a fine-tune you haven't measured is shipping a guess.

Pratik Dhanave · ·7 min read

Scaling Across GPUs

At some point a model doesn't fit on one GPU, or the traffic doesn't, and you have to spread inference across many. The choices — which kind of parallelism, how to place replicas, when to autoscale — are governed by one unforgiving resource (GPU memory) and one expensive one (inter-GPU communication). Get the memory math right and most scaling decisions follow.

At some point a model doesn't fit on one GPU, or the traffic doesn't. The scaling choices are governed by one unforgiving resource (GPU memory) and one expensive one (inter-GPU communication) — get the memory math right and most decisions follow.

Pratik Dhanave · ·5 min read

DSPy in Production

A compiled DSPy program is an artifact — you optimize it once, save it, and serve it — which turns the framework's compile step into an ordinary part of a build pipeline rather than something that runs on every request.

A compiled DSPy program is an artifact — optimize once, save, and serve — which turns compilation into a build step, and makes a model swap a recompile rather than a prompt rewrite.

Pratik Dhanave · ·5 min read

Cost, FinOps, and Scaling the Capability

A system can be reliable, safe, and responsible and still fail — by being uneconomical — and the final move is turning one production system into a repeatable capability many teams can build safely.

A system can be reliable, safe, and responsible and still fail by being uneconomical. Phases 10-11: known unit economics and spend guardrails, then a paved road that lets many teams ship governed AI safely.

Pratik Dhanave · ·6 min read

Serving Engines and PagedAttention

You don't assemble the KV cache, continuous batching, quantization, and speculative decoding by hand — you use a serving engine that has already solved the hard parts. And the idea that ties them together, PagedAttention, is a borrowed operating-systems trick: manage the KV cache like virtual memory, in pages, and the waste that throttled everything disappears.

You don't assemble the KV cache, batching, and quantization by hand — you use a serving engine. And PagedAttention, the idea that ties them together, is a borrowed OS trick: manage the KV cache like virtual memory, in pages.

Pratik Dhanave · ·5 min read

Responsible AI and Human Factors

Responsible-AI principles written in a policy do nothing; the work of this phase is turning fairness, transparency, explainability, and human oversight into features the running system actually has.

Responsible-AI principles in a policy do nothing; this phase turns fairness, transparency, explainability, and human oversight into features the running system actually has. Phase 9.

Pratik Dhanave · ·5 min read

Observability, Reliability, and Incident Response

You cannot operate what you cannot see, and AI systems fail in ways ordinary monitoring misses — quality silently degrades, cost silently climbs, and inputs silently drift — so observability has to watch the things that don't throw exceptions.

AI fails in ways ordinary monitoring misses — quality degrades, cost climbs, inputs drift, all silently. Phase 8: observability that watches the signals that don't throw exceptions, plus runbooks and a kill-switch.

Pratik Dhanave · ·5 min read

Security and Adversarial Robustness

AI systems add attack surface that traditional security does not cover — the model, its prompts, its retrieved context, and its tools are all attackable — and the only way to know you're defended is to threat-model the whole surface and prove it with red-teaming.

AI adds attack surface conventional security misses — model, prompts, retrieved context, and tools are all attackable. Phase 7: threat-model the whole surface and prove it with red-teaming.

Pratik Dhanave · ·6 min read

Batching and Throughput

A single request leaves an expensive GPU almost entirely idle. Batching is how you fill it — and the leap from naive static batching to continuous batching is the single biggest throughput improvement in modern LLM serving, often several times more requests from the exact same hardware.

A single request leaves an expensive GPU almost entirely idle. Batching fills it — and the leap from static to continuous batching is the single biggest throughput improvement in modern LLM serving.

Pratik Dhanave · ·4 min read

MLOps / LLMOps: Making Change Safe and Repeatable

The question this phase answers is simple and unforgiving — can you change the system without breaking it? — and the control you cannot skip is that no ungated change reaches users.

Can you change the system without breaking it? Phase 6: automated pipelines for code, models, and prompts, with eval gates, a registry, and automated rollback — no ungated change reaches users.

Pratik Dhanave · ·6 min read

The Fine-Tuning Spectrum

"Fine-tuning" is not one thing — it's a family of techniques that range from cheaply nudging a model's output format to expensively rebuilding its knowledge base. Confusing them leads to using a sledgehammer for a thumbtack. Knowing where your task sits on the spectrum tells you which technique, how much data, and how much compute you actually need.

'Fine-tuning' is not one thing — it's a family from cheaply nudging output format to expensively rebuilding a model's knowledge. Knowing where your task sits on the spectrum tells you which technique, how much data, and how much compute you actually need.

Pratik Dhanave · ·5 min read

Deployment and Serving: Traffic, SLOs, and Inline Guardrails

A model that scores well in evaluation still has to serve real traffic within a latency budget, isolate tenants, plan for capacity, and enforce safety in the request path — and the guardrails have to be inline, not a filter someone can route around.

A model that scores well still has to serve traffic within a latency budget, isolate tenants, plan capacity, and enforce safety in the request path — inline, not as a filter someone can route around. Phase 5.

Pratik Dhanave · ·6 min read

What Fine-Tuning Is and When to Use It

Fine-tuning is the most misunderstood tool in the LLM toolkit. Reach for it to teach a model new facts and you'll waste weeks and get worse results than a day of RAG. Reach for it to change how a model behaves — its format, tone, or a narrow skill — and nothing else comes close. The whole discipline starts with knowing which problem you actually have.

Fine-tuning is the most misunderstood tool in the LLM toolkit. Reach for it to teach a model facts and you'll waste weeks; reach for it to change how a model behaves — format, tone, a narrow skill — and nothing else comes close. The discipline starts with knowing which problem you have.

Pratik Dhanave · ·5 min read

Evaluation: Making Quality Measurable and Gating

AI systems are non-deterministic, so "it looked good in the demo" is not a quality signal — evaluation is the control system for the entire lifecycle, and if it doesn't gate releases, nothing does.

AI is non-deterministic, so 'it looked good in the demo' is not a quality signal. Phase 4: evaluation as the control system that gates every release — measurable and gating, or it's decoration.

Pratik Dhanave · ·5 min read

Building the System: Compose Before You Train

Modern production AI is rarely "a model" — it is a foundation model wrapped in retrieval, context engineering, tools, and guardrails — and the biggest architectural mistake is reaching for fine-tuning before exhausting the cheaper, more reversible options.

Production AI is rarely a model — it's a composed system, and the biggest mistake is fine-tuning before exhausting cheaper, reversible options. Phase 3: compose before you train.

Pratik Dhanave · ·15 min read

Building an AI Governance Program

The capstone of this series — assembling roles, a use-case inventory, lifecycle gates, and policy-as-code into a right-sized governance program that produces evidence instead of paperwork, without crushing the velocity of a two-person team or failing an enterprise audit.

The capstone: assemble a right-sized governance program — roles and accountability, a model/use-case inventory as the backbone, lifecycle gates, and policy-as-code that blocks deploy when an owner, risk tier, model card, eval pass, or monitor is missing.

Pratik Dhanave · ·5 min read

Data Foundations: The Substrate Everything Depends On

Model quality is bounded by data quality, and the defects you tolerate here — poor lineage, silent drift, unmanaged PII, careless retrieval — resurface downstream as hallucinations, bias, privacy incidents, and un-auditable decisions.

Model quality is bounded by data quality; poor lineage, silent drift, unmanaged PII, and careless retrieval resurface as hallucinations and privacy incidents. Phase 2: the governed data foundation.

Pratik Dhanave · ·5 min read

Governance, Risk, and Compliance Before Real Users

Governance is the one phase whose ordering is non-negotiable: every major framework treats it as a lifecycle function established up front, and retrofitting it after an incident is how you end up with unexplainable models and regulatory exposure.

Governance is the one phase whose ordering is non-negotiable — established before real users, not retrofitted after an incident. Phase 1: NIST AI RMF, EU AI Act risk tiers, ISO 42001, and the artifacts to produce now.

Pratik Dhanave · ·13 min read

Monitoring and Drift in Production

Governance doesn't stop at deploy. This is the NIST RMF MANAGE function in practice: what to monitor for an LLM system, how to detect the drift — including the silent kind where a provider swaps the model under you — and why the audit trail you log is the regulatory deliverable, not a debugging convenience.

Governance doesn't stop at deploy — the Manage function in production: what to monitor (operational, quality, safety), detecting drift including silent provider-model drift behind moving aliases, and the audit trail (versions + retrieved context, PII-redacted) that becomes regulatory evidence.

Pratik Dhanave · ·5 min read

Strategy, Use-Case Selection, and the Named Owner

The most expensive AI failures are systems that work technically but solve the wrong problem, cannot show a return, or have no one accountable when they misbehave — and all three are decided before a single model is chosen.

The most expensive AI failures solve the wrong problem, can't show a return, or have no accountable owner — all decided before a model is chosen. Phase 0: strategy, use-case selection, and the named owner.

Pratik Dhanave · ·5 min read

Why AI Prototypes Don't Reach Production

Organizations rarely fail at building an AI demo; they fail at the gap between a working prototype and a governed, reliable, cost-controlled system — and that gap has a shape you can map.

Organizations rarely fail at building an AI demo; they fail at the gap to a governed, reliable, cost-controlled system. This roadmap maps that gap as twelve phases and a maturity ladder.

All posts on this site are written by Pratik Dhanave, an Agentic AI Architect with 7+ years building production distributed systems, multi-agent AI platforms, and cloud-native infrastructure. About the author → Each article includes working code, architecture diagrams, and references to the specific frameworks and standards discussed. Browse all posts or explore related topics using the tag cloud above.