Start a project
Menu
ServicesAll servicesCybersecurityCloud EngineeringApplication ServicesEnterprise PlatformsData & AnalyticsAI & Machine LearningInternet of ThingsMobilize Talent
PlatformsAll platformsIdentity and accessPrivileged access and secretsERP, HCM and financeCRM, ITSM and automationHealthcareInsurance and bankingData and integrationCloud and AIIndustrial, utilities and mainframeSkills matrix
Case studiesAll case studiesIdentity and accessPrivileged accessSecurity operationsCloud and infrastructureApplicationsERP and HCMCRM, ITSM and automationIndustry platformsData and analyticsAI and agents
IndustriesAll industriesEnergy, oil & gasHealthcare payersBanking & financial servicesInsurancePublic sectorManufacturingLocations
InsightsField notesLearning tracks
CompanyAbout CohashPortfolioSkills matrixCareersTrust and securityContact
Start a project
ai platform

OpenAI integrations that keep working when the model list changes.

We build applications and agents on the OpenAI API and hook them up to your data and identity systems. We also keep an eye on OpenAI's deprecation schedule, because a retired model shouldn't be how you find out production is broken.

Talk to a OpenAI API engineerWhat changed recently
platform dossierOpenAI APIAI platform
Vendor
OpenAI
Latest models
GPT-6 Astra (3 Sep 2026), GPT-6.1 Sol (29 Sep 2026)
Next shutdown
GPT-3.5, o1 and o3-mini on 23 Oct 2026
Then
v1/prompts API and Agent Builder on 30 Nov 2026
release watch
  • GPT-6 Astra released03 Sep 2026
  • Agents API in public beta10 Sep 2026
connects with
in short·20-second read

Cohash builds production applications on the OpenAI API.

We use the Responses API and the new Agents API, connect models to your data through tools and retrieval, add evaluation and logging, and control keys, projects and spend. OpenAI retires models often. The Assistants API shut down on 26 August 2026, and older models such as o1 and GPT-3.5 shut down on 23 October 2026, so we plan and test migrations ahead of each date.

you probably need this if
  • Your code still calls a model that appears on OpenAI's deprecations page.
  • An internal tool still runs on the Assistants API and nobody has moved it to Responses yet.
  • One API key is shared by several teams and nobody has set a spend limit.
Ask an engineer facts checked 2026-10-20 · 3 sources
how it runs
  1. 01Usage and model inventory
  2. 02Evaluation set first
  3. 03Migrate and add tools
  4. 04Projects, keys and limits
  5. 05Monitor and hand over
you keep
  • Model ID inventory with shutdown dates taken from OpenAI's deprecations page
  • Evaluation set with automated scoring on every model or prompt change
  • Migration plan for saved prompt objects ahead of the v1/prompts shutdown
+ 2 more documents
who does it
AI engineer (OpenAI API, Agents SDK)LLM evaluation engineerRetrieval and RAG engineer
platforms
proof
4 case studies on this
background
01 Telemetry feedback loop failures feed the golden set 02 Citations and refusals cite chunks, refuse guesses 03 Eval harness in CI blocks merges on regression 04 Hybrid retrieval BM25, dense, reranker 05 ACL-aware ingestion re-embed on update, tombstones
field note · 9 minRAG in production: what actually breaks
openai-api · release timeline
  1. 26 Aug 2026 Assistants API shut down Shut down 26 August 2026. OpenAI recommends the Responses and Conversations APIs. source ↗
  2. 03 Sep 2026 GPT-6 Astra released OpenAI's most capable API model at release, followed by GPT-6 Sol and GPT-6 Luna on 22 September and GPT-6.1 Sol on 29 September 2026. source ↗
  3. 10 Sep 2026 Agents API in public beta Released 10 September 2026. Computer use added to the Agents API on 29 September 2026. source ↗
  4. today · Oct 22, 2026
  5. no date yetUpcoming shutdownsLegacy models including GPT-3.5, o1 and o3-mini shut down 23 October 2026. The v1/prompts API and Agent Builder shut down 30 November 2026.source ↗
the story · openai api estate

From the first call to a setup your own team can run.

Here's what usually prompts a call about your OpenAI API estate, what we do about it and in which order, and where you end up.

  1. chapter 01 · where it hurts

    Why teams reach out

    • Your code still calls a model that appears on OpenAI's deprecations page.
    • An internal tool still runs on the Assistants API and nobody has moved it to Responses yet.
    • One API key is shared by several teams and nobody has set a spend limit.
  2. chapter 02 · what we change

    How we solve it

    • Model and API migrations
    • Agents and tool use
    • Retrieval on your data
    • Evaluation and monitoring
    • Security, keys and spend
  3. chapter 03 · how it runs

    Our approach

    • Usage and model inventory
    • Evaluation set first
    • Migrate and add tools
    • Projects, keys and limits
    • Monitor and hand over
  4. chapter 04 · where you end up

    What changes

    • The batch job and the app both run on current models, and the scores that picked them are written down.
    • Teams see their own usage against their own limit.
    • Model ID inventory with shutdown dates taken from OpenAI's deprecations page
    • Evaluation set with automated scoring on every model or prompt change
what we do

Our OpenAI API work

5 kinds of OpenAI API work, each run by engineers who have done it on live tenants.

  1. Code moved off retired models and the Assistants API onto the Responses and Conversations APIs, with side-by-side evaluation before cut-over.

  2. Agents built with the Agents API or Agents SDK, calling your internal APIs through tools with scoped credentials and logged actions.

  3. File search or your own vector store over approved documents, with access checks so users only get answers from content they may already see.

  4. Test sets built from real questions, automated scoring on every prompt or model change, and production logging for cost, latency and failure review.

  5. Projects, service accounts and key rotation set up per application, with hard spend limits and data retention settings reviewed with your security team.

capability 01 / 05in production

Model and API migrations

Code moved off retired models and the Assistants API onto the Responses and Conversations APIs, with side-by-side evaluation before cut-over.

talk to an engineer about this →
capability 02 / 05in production

Agents and tool use

Agents built with the Agents API or Agents SDK, calling your internal APIs through tools with scoped credentials and logged actions.

talk to an engineer about this →
capability 03 / 05in production

Retrieval on your data

File search or your own vector store over approved documents, with access checks so users only get answers from content they may already see.

talk to an engineer about this →
capability 04 / 05in production

Evaluation and monitoring

Test sets built from real questions, automated scoring on every prompt or model change, and production logging for cost, latency and failure review.

talk to an engineer about this →
capability 05 / 05in production

Security, keys and spend

Projects, service accounts and key rotation set up per application, with hard spend limits and data retention settings reviewed with your security team.

talk to an engineer about this →
typical situations

OpenAI API situations we walk into

We pieced these together from OpenAI API situations that come up a lot. They're patterns. You won't find a specific client in them.

case file 01

A nightly batch job still pinned to a model with a shutdown date

composite · not a client story
  1. Beforethe situation

    A nightly claims-summary job is pinned to o1-2024-12-17, and a customer-facing app depends on saved prompt objects through v1/prompts. Both have published shutdown dates. The only testing anyone does is one person reading ten outputs.

  2. What we dothe work

    We pull a few hundred past inputs into a graded test set and run candidate GPT-6 models against it, choosing on score, cost and latency. Prompt objects move into versioned prompts in your own repository, and model IDs become configuration so the next swap is a change request.

  3. Afterwhere you end up

    The batch job and the app both run on current models, and the scores that picked them are written down. Prompts sit in your repo next to the code. When the next shutdown date is posted, someone reruns the test set.

case file 02

One API key shared by every team in the company

composite · not a client story
  1. Beforethe situation

    Every internal app reads the same OpenAI API key from one vault path. It has never been rotated. Finance can't split the bill by team, and security has never looked at the retention settings.

  2. What we dothe work

    We create a project for each application, with its own service account and key, and set a spend limit on each. Keys rotate through your secrets manager. Security reviews the data retention and logging settings with us, and we record what was agreed.

  3. Afterwhere you end up

    Teams see their own usage against their own limit. If a key leaks, one app is affected and one key gets revoked. Security has the retention settings in writing.

case studies

OpenAI API case studies

All case studies
upgrade AI and agents

OpenAI Assistants API to the Responses API and the Agents SDK

Getting apps off the shut-down OpenAI Assistants API onto the Responses and Conversations APIs or the Agents SDK, with history kept and outputs re-tested.

read the case study
AI rollout AI and agents

AI pilot to production with forward deployed engineers

Taking an AI pilot into production with forward deployed engineers who work with the business team, build evals, connect real systems and hand over cleanly.

read the case study
AI rollout AI and agents

Enterprise search and RAG that respects document permissions

Enterprise search and RAG that only return what each user may open. We capture source ACLs at ingestion, filter at query time and leak-test.

read the case study
approach

How a OpenAI API engagement runs

We don't start a step until the last one has produced something you can check.

  1. step 01

    Usage and model inventory

    We find every model, endpoint and key your applications call, and match them against OpenAI's deprecation schedule.

  2. step 02

    Evaluation set first

    Before anybody touches a prompt or a model, we build a test set from real requests and write scoring rules. Changes then win or lose on evidence.

  3. step 03

    Migrate and add tools

    Code moves to the Responses API or Agents API, and tools call your internal APIs through your own service with scoped credentials.

  4. step 04

    Projects, keys and limits

    Each application gets its own project, service account and spend limit, with retention settings reviewed with your security team.

  5. step 05

    Monitor and hand over

    Logging for cost, latency and failures goes live with the application, and your team keeps a routine for checking the deprecation list.

ai on openai api

Where AI fits on OpenAI API, and where it needs guardrails.

From OpenAI's API you get models and plumbing. The plumbing is the Responses API, the Agents API (in public beta since 10 September 2026) and a remote MCP tool so models can call MCP servers. Things retire often here. Agent Builder and the v1/prompts API shut down on 30 November 2026. Expect the engineering to be version tracking, evaluation and tool security.

AI agentasks for data or an action
Scoped identityonly the access it needs
MCP serverthe one door into the platform
Logged, approveda person signs off on writes
OpenAI APIdoes the work

In this setup the MCP client is OpenAI. The Responses API's mcp tool calls the servers you point it at, with OAuth tokens in the authorization field per request, never stored. We put the MCP servers behind your identity provider and keep require_approval on for writes. We also restrict allowed_tools and log every mcp_call item.

Responses API and GPT-6 models

OpenAI released GPT-6 Astra on 3 September 2026 as its most capable API model at the time. GPT-6 Sol and GPT-6 Luna came later that month, as did GPT-6.1 Sol. We keep model IDs in configuration, and a candidate model has to pass a graded test set before we switch.

Agents API

The Agents API runs multi-step agents, and on 29 September 2026 it picked up computer use. It's still in public beta. So we put it behind a flag on production paths and keep a Responses API fallback.

Remote MCP tool

Public MCP servers are reached by URL from the Responses API, and private ones through Secure MCP Tunnel. Out of the box it asks for approval before it sends data to a server, and allowed_tools controls which tools get imported. We leave approval on for every tool that writes.

Deprecations to plan for

Three changes to watch. The Assistants API shut down on 26 August 2026. Legacy models such as o1 and GPT-3.5 shut down on 23 October. And for models released after 1 September 2026, connectors set by connector_id are deprecated. We check the deprecation list against your code once a month.

OpenAI API: MCP and connectors guide ↗OpenAI API changelog ↗OpenAI API deprecations ↗
architecture

How an OpenAI-backed application reaches your data and APIs

Users sign in through Entra ID, and your service sends requests to the Responses or Agents API through a gateway that holds keys and limits. When the model asks for a tool, your service calls internal APIs with scoped credentials and logs the result.

How an OpenAI-backed application reaches your data and APIs Users sign in through Entra ID, and your service sends requests to the Responses or Agents API through a gateway that holds keys and limits. When the model asks for a tool, your service calls internal APIs with scoped credentials and logs the result. 01 · APPLICATION 02 · YOUR SERVICE LAYER 03 · OPENAI 04 · YOUR DATA AND SYSTEMS OIDC sign-in request API call agent run file search function calls scoped queries eval runs Ex Entra ID user sign-in ·groups Your application web app · Teams · portal Model gateway keys · spend limits · logs Evaluation suite test sets · graders AI Responses API GPT-6 · functioncalling Agents API multi-step agent runs Vector store approved documents Internal APIs orders · tickets · HR Snowflake read-only analyticsrole
Read the diagram as text
  1. Entra ID → Your application: OIDC sign-in
  2. Your application → Model gateway: request
  3. Model gateway → Responses API: API call
  4. Model gateway → Agents API: agent run
  5. Responses API → Vector store: file search
  6. Model gateway → Internal APIs: function calls
  7. Model gateway → Snowflake: scoped queries
  8. Evaluation suite → Responses API: eval runs
use cases

Typical OpenAI API projects

The work teams usually bring us, and the platforms OpenAI API has to talk to while we do it.

  1. OPE-01 Your application still calls a model on OpenAI's deprecation list common request
  2. OPE-02 You built on the Assistants API and need to move to Responses common request
  3. OPE-03 A chatbot answers from documents some users shouldn't see common request
  4. OPE-04 Nobody can say whether a prompt change made answers better or worse common request
  5. OPE-05 API keys are shared across teams with no spend limits common request
OpenAI API · connected platforms
Anthropic ClaudeMicrosoft Entra IDMicrosoft Power PlatformDatabricksSnowflake
you keep

What a OpenAI API engagement leaves behind

These are working documents, written against your configuration, so your team can run them without us.

  • inventory01

    Model ID inventory with shutdown dates taken from OpenAI's deprecations page

    yours to keep
  • design02

    Evaluation set with automated scoring on every model or prompt change

    yours to keep
  • plan03

    Migration plan for saved prompt objects ahead of the v1/prompts shutdown

    yours to keep
  • document04

    Project and service account layout with key rotation and spend limits

    yours to keep
  • design05

    Retrieval design with access checks for each document source

    yours to keep
talent and delivery · openai api

Certified engineers, wherever you need them.

Some clients want a project team, some want a couple of engineers on theirs. Either works. Onsite, nearshore or offshore, with handovers across time zones that keep things moving.

Onsite

Engineers at your sites in the US

Nearshore

Overlapping hours with US teams

Offshore

Overnight progress on your backlog

All time zones

Coverage handed over across time zones

24-hour coverage · US Central time
OnsiteUS business hours
Nearshoreoverlapping US hours
Offshoreovernight progress
Combinedhandover, not hand-off
roles we staff
  • AI engineer (OpenAI API, Agents SDK)
  • LLM evaluation engineer
  • Retrieval and RAG engineer
  • AI application back-end developer
  • AI security engineer
  • Forward deployed AI engineer
due diligence · openai api

Questions to ask any partner, including us.

Put these to anyone bidding on your OpenAI API work. The ones who've done it answer differently from the ones who've read about it.

  1. Q1

    How are you going to keep OpenAI's deprecation dates matched against the model IDs sitting in our code?

    OpenAI lists shutdowns most months in 2026, including legacy models on 23 October and the v1/prompts API on 30 November. A scheduled check against your own inventory keeps those dates from becoming incidents.

  2. Q2

    Will you pin dated model versions, and how will we move off them later?

    A pinned version does the same thing every day right up until it's shut down. Pin it, and keep an evaluation set ready, so the next version is a test run and nobody has to rewrite anything.

  3. Q3

    Should this build use the Responses API, the Agents API or the Agents SDK?

    The Agents API is in public beta, and Agent Builder shuts down on 30 November 2026. Build production paths on the stable APIs and put beta features behind a flag you can turn off.

  4. Q4

    How will retrieval respect the permissions users already have on our documents?

    Build a vector store from a SharePoint export and the access controls are gone. Unless something checks permissions at query time, users get answers drawn from documents they couldn't open.

  5. Q5

    Should we call OpenAI directly or through Azure, given our contracts?

    New models don't always land on both on the same day, and availability varies. What usually decides it, in our experience, is an existing Azure agreement, a data residency rule or Entra ID integration.

01How often do we need to change models?

Often. Look at OpenAI's deprecations page for 2026 and nearly every month has a shutdown on it. We pin model versions, keep an evaluation set ready and review the list on a schedule, so migrations are planned rather than forced.

02Should we use the OpenAI API directly or through Azure?

It depends on your contracts, data residency needs and which models you require. Azure OpenAI fits organizations already standardized on Azure and Entra ID. Model availability and release timing can differ between the two, so we check both for your use case.

03What drives the cost of an OpenAI-based application?

Mostly the model you choose, how many tokens each request uses and how often you call it. Prompt caching, smaller models for simple steps and shorter context cut cost. We measure real traffic before recommending changes.

04Do you also work with Anthropic Claude?

Yes. Many clients use more than one provider. We design applications so models can be swapped behind a common interface and evaluated on the same tests.

05Can you help us move off Agent Builder before it shuts down?

Yes. OpenAI lists Agent Builder for shutdown on 30 November 2026. We inventory the workflows built in it, rebuild them in your own code on a supported API, and test the new versions against the same inputs before switching.

services

Related services

Projects on OpenAI API usually end up touching one of these, so they're worth a look.

you are hereOpenAI API
sources 3 sources linked last checked 2026-10-20 content fingerprint c80a2a67
  1. 01OpenAI API changelogdevelopers.openai.com ↗
  2. 02OpenAI API deprecationsdevelopers.openai.com ↗
  3. 03OpenAI: Assistants migration guidedevelopers.openai.com ↗
Reviewed by the Cohash team OpenAI API is a trademark of its owner, used here to identify the platform. how we handle facts and data →
next step

Tell us what you're running. We'll tell you what we'd do first.

A 30-minute working session with a senior engineer. No slides.

Book a working session See the case studies
useful to bring
  • The platform and the version you're on
  • Where you need to get to, and by when
  • What's custom: rules, plugins, integrations
  • Any audit finding or deadline driving it